kallithea Changeset - d7488551578e

Changeset - d7488551578e

Parent rev.

Child rev.

[Not reviewed]

beta

1 58 1

Marcin Kuzminski - 13 years ago 2013-04-27 11:24:25
marcin@python-works.com

synced vcs with upstream

- moved subprocessio module to VCS
- many small changes to make embedded vcs as similar to to external lib
- use only absolute imports
- patch vcs config during load pylons env

18 files changed:

rhodecode/__init__.py

rhodecode/config/environment.py

rhodecode/lib/__init__.py

rhodecode/lib/compat.py

173

rhodecode/lib/middleware/pygrack.py

rhodecode/lib/subprocessio.py

415

rhodecode/lib/utils.py

rhodecode/lib/vcs/__init__.py

rhodecode/lib/vcs/backends/base.py

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

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

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

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

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

rhodecode/lib/vcs/conf/settings.py

rhodecode/lib/vcs/nodes.py

rhodecode/lib/vcs/subprocessio.py

415

rhodecode/lib/vcs/utils/__init__.py

Changeset was too big and was cut off... Show full diff anyway

0 comments (0 inline, 0 general)

rhodecode/__init__.py

➞

Show inline comments

 # -*- coding: utf-8 -*-
 """
     rhodecode.__init__
     ~~~~~~~~~~~~~~~~~~
     RhodeCode, a web based repository management based on pylons
     versioning implementation: http://www.python.org/dev/peps/pep-0386/
     :created_on: Apr 9, 2010
     :author: marcink
     :copyright: (C) 2010-2012 Marcin Kuzminski <marcin@python-works.com>
     :license: GPLv3, see COPYING for more details.
 """
 # This program is free software: you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by
 # the Free Software Foundation, either version 3 of the License, or
 # (at your option) any later version.
+#
 # This program is distributed in the hope that it will be useful,
 # but WITHOUT ANY WARRANTY; without even the implied warranty of
 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 # GNU General Public License for more details.
+#
 # You should have received a copy of the GNU General Public License
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 import sys
 import platform
 VERSION = (1, 7, 0, 'dev')
 BACKENDS = {
     'hg': 'Mercurial repository',
     'git': 'Git repository',
+}
 CELERY_ON = False
 CELERY_EAGER = False
 # link to config for pylons
 CONFIG = {}
 # Linked module for extensions
 EXTENSIONS = {}
 try:
     from rhodecode.lib import get_current_revision
     _rev = get_current_revision()
     if _rev and len(VERSION) > 3:
         VERSION += ('%s' % _rev[0],)
 except ImportError:
     pass
 __version__ = ('.'.join((str(each) for each in VERSION[:3])) +
                '.'.join(VERSION[3:]))
 __dbversion__ = 12  # defines current db version for migrations
 __platform__ = platform.system()
 __license__ = 'GPLv3'
 __py_version__ = sys.version_info
 __author__ = 'Marcin Kuzminski'
 __url__ = 'http://rhodecode.org'
 PLATFORM_WIN = ('Windows')
 PLATFORM_OTHERS = ('Linux', 'Darwin', 'FreeBSD', 'OpenBSD', 'SunOS') #depracated
 is_windows = __platform__ in PLATFORM_WIN
 is_windows = __platform__ in ('Windows')
 is_unix = not is_windows
 BACKENDS = {
     'hg': 'Mercurial repository',
     'git': 'Git repository',
+}
 CELERY_ON = False
 CELERY_EAGER = False
 # link to config for pylons
 CONFIG = {}
 # Linked module for extensions
 EXTENSIONS = {}

rhodecode/config/environment.py

➞

Show inline comments

 """Pylons environment configuration"""
 import os
 import logging
 import rhodecode
 from mako.lookup import TemplateLookup
 from pylons.configuration import PylonsConfig
 from pylons.error import handle_mako_error
 # don't remove this import it does magic for celery
 from rhodecode.lib import celerypylons
 import rhodecode.lib.app_globals as app_globals
 from rhodecode.config.routing import make_map
 from rhodecode.lib import helpers
 from rhodecode.lib.auth import set_available_permissions
 from rhodecode.lib.utils import repo2db_mapper, make_ui, set_rhodecode_config,\
     load_rcextensions, check_git_version
+    load_rcextensions, check_git_version, set_vcs_config
 from rhodecode.lib.utils2 import engine_from_config, str2bool
 from rhodecode.lib.db_manage import DbManage
 from rhodecode.model import init_model
 from rhodecode.model.scm import ScmModel
 log = logging.getLogger(__name__)
 def load_environment(global_conf, app_conf, initial=False):
     """
     Configure the Pylons environment via the ``pylons.config``
     object
     """
     config = PylonsConfig()
     # Pylons paths
     root = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
     paths = dict(
         root=root,
         controllers=os.path.join(root, 'controllers'),
         static_files=os.path.join(root, 'public'),
         templates=[os.path.join(root, 'templates')]
+    )
     # Initialize config with the basic options
     config.init_app(global_conf, app_conf, package='rhodecode', paths=paths)
     # store some globals into rhodecode
     rhodecode.CELERY_ON = str2bool(config['app_conf'].get('use_celery'))
     rhodecode.CELERY_EAGER = str2bool(config['app_conf'].get('celery.always.eager'))
     config['routes.map'] = make_map(config)
     config['pylons.app_globals'] = app_globals.Globals(config)
     config['pylons.h'] = helpers
     rhodecode.CONFIG = config
     load_rcextensions(root_path=config['here'])
     # Setup cache object as early as possible
     import pylons
     pylons.cache._push_object(config['pylons.app_globals'].cache)
     # Create the Mako TemplateLookup, with the default auto-escaping
     config['pylons.app_globals'].mako_lookup = TemplateLookup(
         directories=paths['templates'],
         error_handler=handle_mako_error,
         module_directory=os.path.join(app_conf['cache_dir'], 'templates'),
         input_encoding='utf-8', default_filters=['escape'],
         imports=['from webhelpers.html import escape'])
     # sets the c attribute access when don't existing attribute are accessed
     config['pylons.strict_tmpl_context'] = True
     test = os.path.split(config['__file__'])[-1] == 'test.ini'
     if test:
         if os.environ.get('TEST_DB'):
             # swap config if we pass enviroment variable
             config['sqlalchemy.db1.url'] = os.environ.get('TEST_DB')
         from rhodecode.lib.utils import create_test_env, create_test_index
         from rhodecode.tests import  TESTS_TMP_PATH
         # set RC_NO_TMP_PATH=1 to disable re-creating the database and
         # test repos
         if not int(os.environ.get('RC_NO_TMP_PATH', 0)):
             create_test_env(TESTS_TMP_PATH, config)
         # set RC_WHOOSH_TEST_DISABLE=1 to disable whoosh index during tests
         if not int(os.environ.get('RC_WHOOSH_TEST_DISABLE', 0)):
             create_test_index(TESTS_TMP_PATH, config, True)
     #check git version
     check_git_version()
     DbManage.check_waitress()
     # MULTIPLE DB configs
     # Setup the SQLAlchemy database engine
     sa_engine_db1 = engine_from_config(config, 'sqlalchemy.db1.')
     init_model(sa_engine_db1)
     repos_path = make_ui('db').configitems('paths')[0][1]
     if str2bool(config.get('initial_repo_scan', True)):
         repo2db_mapper(ScmModel().repo_scan(repos_path),
                        remove_obsolete=False, install_git_hook=False)
     set_available_permissions(config)
     config['base_path'] = repos_path
     set_rhodecode_config(config)
     instance_id = rhodecode.CONFIG.get('instance_id')
     if instance_id == '*':
         instance_id = '%s-%s' % (os.uname()[1], os.getpid())
         rhodecode.CONFIG['instance_id'] = instance_id
     # CONFIGURATION OPTIONS HERE (note: all config options will override
     # any Pylons config options)
     # store config reference into our module to skip import magic of
     # pylons
     rhodecode.CONFIG.update(config)
     set_vcs_config(rhodecode.CONFIG)
     return config

rhodecode/lib/__init__.py

➞

Show inline comments

 import os
 def get_current_revision(quiet=False):
     """
     Returns tuple of (number, id) from repository containing this package
     or None if repository could not be found.
     :param quiet: prints error for fetching revision if True
     """
     try:
         from rhodecode.lib.vcs import get_repo
         from rhodecode.lib.vcs.utils.helpers import get_scm
         repopath = os.path.join(os.path.dirname(__file__), '..', '..')
         repopath = os.path.abspath(os.path.join(os.path.dirname(__file__),
                                                 '..', '..'))
         scm = get_scm(repopath)[0]
         repo = get_repo(path=repopath, alias=scm)
         wk_dir = repo.workdir
         cur_rev = wk_dir.get_changeset()
         return (cur_rev.revision, cur_rev.short_id)
     except Exception, err:
         if not quiet:
             print ("WARNING: Cannot retrieve rhodecode's revision. "
                    "disregard this if you don't know what that means. "
                    "Original error was: %s" % err)
         return None

rhodecode/lib/compat.py

➞

Show inline comments

 # -*- coding: utf-8 -*-
 """
     rhodecode.lib.compat
     ~~~~~~~~~~~~~~~~~~~~
     Python backward compatibility functions and common libs
     :created_on: Oct 7, 2011
     :author: marcink
     :copyright: (C) 2010-2010 Marcin Kuzminski <marcin@python-works.com>
     :license: GPLv3, see COPYING for more details.
 """
 # This program is free software: you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by
 # the Free Software Foundation, either version 3 of the License, or
 # (at your option) any later version.
+#
 # This program is distributed in the hope that it will be useful,
 # but WITHOUT ANY WARRANTY; without even the implied warranty of
 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 # GNU General Public License for more details.
+#
 # You should have received a copy of the GNU General Public License
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 import os
-from rhodecode import __platform__, PLATFORM_WIN, __py_version__
+from rhodecode import __py_version__, is_windows
 #==============================================================================
 # json
 #==============================================================================
 from rhodecode.lib.ext_json import json
 import array
 #==============================================================================
 # izip_longest
 #==============================================================================
 try:
     from itertools import izip_longest
 except ImportError:
     import itertools
     def izip_longest(*args, **kwds):
         fillvalue = kwds.get("fillvalue")
         def sentinel(counter=([fillvalue] * (len(args) - 1)).pop):
             yield counter()  # yields the fillvalue, or raises IndexError
         fillers = itertools.repeat(fillvalue)
         iters = [itertools.chain(it, sentinel(), fillers)
                     for it in args]
         try:
             for tup in itertools.izip(*iters):
                 yield tup
         except IndexError:
             pass
 #==============================================================================
 # OrderedDict
 #==============================================================================
 # Python Software Foundation License
 # XXX: it feels like using the class with "is" and "is not" instead of "==" and
 # "!=" should be faster.
 class _Nil(object):
     def __repr__(self):
         return "nil"
     def __eq__(self, other):
         if (isinstance(other, _Nil)):
             return True
         else:
             return NotImplemented
     def __ne__(self, other):
         if (isinstance(other, _Nil)):
             return False
         else:
             return NotImplemented
 _nil = _Nil()
 class _odict(object):
     """Ordered dict data structure, with O(1) complexity for dict operations
     that modify one element.
     Overwriting values doesn't change their original sequential order.
     """
     def _dict_impl(self):
         return None
     def __init__(self, data=(), **kwds):
         """This doesn't accept keyword initialization as normal dicts to avoid
         a trap - inside a function or method the keyword args are accessible
         only as a dict, without a defined order, so their original order is
         lost.
         """
         if kwds:
             raise TypeError("__init__() of ordered dict takes no keyword "
                             "arguments to avoid an ordering trap.")
         self._dict_impl().__init__(self)
         # If you give a normal dict, then the order of elements is undefined
         if hasattr(data, "iteritems"):
             for key, val in data.iteritems():
                 self[key] = val
         else:
             for key, val in data:
                 self[key] = val
     # Double-linked list header
     def _get_lh(self):
         dict_impl = self._dict_impl()
         if not hasattr(self, '_lh'):
             dict_impl.__setattr__(self, '_lh', _nil)
         return dict_impl.__getattribute__(self, '_lh')
     def _set_lh(self, val):
         self._dict_impl().__setattr__(self, '_lh', val)
     lh = property(_get_lh, _set_lh)
     # Double-linked list tail
     def _get_lt(self):
         dict_impl = self._dict_impl()
         if not hasattr(self, '_lt'):
             dict_impl.__setattr__(self, '_lt', _nil)
         return dict_impl.__getattribute__(self, '_lt')
     def _set_lt(self, val):
         self._dict_impl().__setattr__(self, '_lt', val)
     lt = property(_get_lt, _set_lt)
     def __getitem__(self, key):
         return self._dict_impl().__getitem__(self, key)[1]
     def __setitem__(self, key, val):
         dict_impl = self._dict_impl()
         try:
             dict_impl.__getitem__(self, key)[1] = val
         except KeyError:
             new = [dict_impl.__getattribute__(self, 'lt'), val, _nil]
             dict_impl.__setitem__(self, key, new)
             if dict_impl.__getattribute__(self, 'lt') == _nil:
                 dict_impl.__setattr__(self, 'lh', key)
             else:
                 dict_impl.__getitem__(
                     self, dict_impl.__getattribute__(self, 'lt'))[2] = key
             dict_impl.__setattr__(self, 'lt', key)
     def __delitem__(self, key):
         dict_impl = self._dict_impl()
         pred, _, succ = self._dict_impl().__getitem__(self, key)
         if pred == _nil:
             dict_impl.__setattr__(self, 'lh', succ)
         else:
             dict_impl.__getitem__(self, pred)[2] = succ
         if succ == _nil:
             dict_impl.__setattr__(self, 'lt', pred)
         else:
             dict_impl.__getitem__(self, succ)[0] = pred
         dict_impl.__delitem__(self, key)
     def __contains__(self, key):
         return key in self.keys()
     def __len__(self):
         return len(self.keys())
     def __str__(self):
         pairs = ("%r: %r" % (k, v) for k, v in self.iteritems())
         return "{%s}" % ", ".join(pairs)
     def __repr__(self):
         if self:
             pairs = ("(%r, %r)" % (k, v) for k, v in self.iteritems())
             return "odict([%s])" % ", ".join(pairs)
         else:
             return "odict()"
     def get(self, k, x=None):
         if k in self:
             return self._dict_impl().__getitem__(self, k)[1]
         else:
             return x
     def __iter__(self):
         dict_impl = self._dict_impl()
         curr_key = dict_impl.__getattribute__(self, 'lh')
         while curr_key != _nil:
             yield curr_key
             curr_key = dict_impl.__getitem__(self, curr_key)[2]
     iterkeys = __iter__
     def keys(self):
         return list(self.iterkeys())
     def itervalues(self):
         dict_impl = self._dict_impl()
         curr_key = dict_impl.__getattribute__(self, 'lh')
         while curr_key != _nil:
             _, val, curr_key = dict_impl.__getitem__(self, curr_key)
             yield val
     def values(self):
         return list(self.itervalues())
     def iteritems(self):
         dict_impl = self._dict_impl()
         curr_key = dict_impl.__getattribute__(self, 'lh')
         while curr_key != _nil:
             _, val, next_key = dict_impl.__getitem__(self, curr_key)
             yield curr_key, val
             curr_key = next_key
     def items(self):
         return list(self.iteritems())
     def sort(self, cmp=None, key=None, reverse=False):
         items = [(k, v) for k, v in self.items()]
         if cmp is not None:
             items = sorted(items, cmp=cmp)
         elif key is not None:
             items = sorted(items, key=key)
         else:
             items = sorted(items, key=lambda x: x[1])
         if reverse:
             items.reverse()
         self.clear()
         self.__init__(items)
     def clear(self):
         dict_impl = self._dict_impl()
         dict_impl.clear(self)
         dict_impl.__setattr__(self, 'lh', _nil)
         dict_impl.__setattr__(self, 'lt', _nil)
     def copy(self):
         return self.__class__(self)
     def update(self, data=(), **kwds):
         if kwds:
             raise TypeError("update() of ordered dict takes no keyword "
                             "arguments to avoid an ordering trap.")
         if hasattr(data, "iteritems"):
             data = data.iteritems()
         for key, val in data:
             self[key] = val
     def setdefault(self, k, x=None):
         try:
             return self[k]
         except KeyError:
             self[k] = x
             return x
     def pop(self, k, x=_nil):
         try:
             val = self[k]
             del self[k]
             return val
         except KeyError:
             if x == _nil:
                 raise
             return x
     def popitem(self):
         try:
             dict_impl = self._dict_impl()
             key = dict_impl.__getattribute__(self, 'lt')
             return key, self.pop(key)
         except KeyError:
             raise KeyError("'popitem(): ordered dictionary is empty'")
     def riterkeys(self):
         """To iterate on keys in reversed order.
         """
         dict_impl = self._dict_impl()
         curr_key = dict_impl.__getattribute__(self, 'lt')
         while curr_key != _nil:
             yield curr_key
             curr_key = dict_impl.__getitem__(self, curr_key)[0]
     __reversed__ = riterkeys
     def rkeys(self):
         """List of the keys in reversed order.
         """
         return list(self.riterkeys())
     def ritervalues(self):
         """To iterate on values in reversed order.
         """
         dict_impl = self._dict_impl()
         curr_key = dict_impl.__getattribute__(self, 'lt')
         while curr_key != _nil:
             curr_key, val, _ = dict_impl.__getitem__(self, curr_key)
             yield val
     def rvalues(self):
         """List of the values in reversed order.
         """
         return list(self.ritervalues())
     def riteritems(self):
         """To iterate on (key, value) in reversed order.
         """
         dict_impl = self._dict_impl()
         curr_key = dict_impl.__getattribute__(self, 'lt')
         while curr_key != _nil:
             pred_key, val, _ = dict_impl.__getitem__(self, curr_key)
             yield curr_key, val
             curr_key = pred_key
     def ritems(self):
         """List of the (key, value) in reversed order.
         """
         return list(self.riteritems())
     def firstkey(self):
         if self:
             return self._dict_impl().__getattribute__(self, 'lh')
         else:
             raise KeyError("'firstkey(): ordered dictionary is empty'")
     def lastkey(self):
         if self:
             return self._dict_impl().__getattribute__(self, 'lt')
         else:
             raise KeyError("'lastkey(): ordered dictionary is empty'")
     def as_dict(self):
         return self._dict_impl()(self.items())
     def _repr(self):
         """_repr(): low level repr of the whole data contained in the odict.
         Useful for debugging.
         """
         dict_impl = self._dict_impl()
         form = "odict low level repr lh,lt,data: %r, %r, %s"
         return form % (dict_impl.__getattribute__(self, 'lh'),
                        dict_impl.__getattribute__(self, 'lt'),
                        dict_impl.__repr__(self))
 class OrderedDict(_odict, dict):
     def _dict_impl(self):
         return dict
 #==============================================================================
 # OrderedSet
 #==============================================================================
 from sqlalchemy.util import OrderedSet
 #==============================================================================
 # kill FUNCTIONS
 #==============================================================================
-if __platform__ in PLATFORM_WIN:
+if is_windows:
     import ctypes
     def kill(pid, sig):
         """kill function for Win32"""
         kernel32 = ctypes.windll.kernel32
         handle = kernel32.OpenProcess(1, 0, pid)
         return (0 != kernel32.TerminateProcess(handle, 0))
 else:
     kill = os.kill
 #==============================================================================
 # itertools.product
 #==============================================================================
 try:
     from itertools import product
 except ImportError:
     def product(*args, **kwds):
         # product('ABCD', 'xy') --> Ax Ay Bx By Cx Cy Dx Dy
         # product(range(2), repeat=3) --> 000 001 010 011 100 101 110 111
         pools = map(tuple, args) * kwds.get('repeat', 1)
         result = [[]]
         for pool in pools:
             result = [x + [y] for x in result for y in pool]
         for prod in result:
             yield tuple(prod)
 #==============================================================================
 # BytesIO
 #==============================================================================
 try:
     from io import BytesIO
 except ImportError:
     from cStringIO import StringIO as BytesIO
 #==============================================================================
 # bytes
 #==============================================================================
 if __py_version__ >= (2, 6):
     _bytes = bytes
 else:
     # in py2.6 bytes is a synonim for str
     _bytes = str
 if __py_version__ >= (2, 6):
     _bytearray = bytearray
 else:
     import array
     # no idea if this is correct but all integration tests are passing
     # i think we never use bytearray anyway
     _bytearray = array
 #==============================================================================
 # deque
 #==============================================================================
 if __py_version__ >= (2, 6):
     from collections import deque
 else:
     #need to implement our own deque with maxlen
     class deque(object):
         def __init__(self, iterable=(), maxlen= -1):
             if not hasattr(self, 'data'):
                 self.left = self.right = 0
                 self.data = {}
             self.maxlen = maxlen or -1
             self.extend(iterable)
         def append(self, x):
             self.data[self.right] = x
             self.right += 1
             if self.maxlen != -1 and len(self) > self.maxlen:
                 self.popleft()
         def appendleft(self, x):
             self.left -= 1
             self.data[self.left] = x
             if self.maxlen != -1 and len(self) > self.maxlen:
                 self.pop()
         def pop(self):
             if self.left == self.right:
                 raise IndexError('cannot pop from empty deque')
             self.right -= 1
             elem = self.data[self.right]
             del self.data[self.right]
             return elem
         def popleft(self):
             if self.left == self.right:
                 raise IndexError('cannot pop from empty deque')
             elem = self.data[self.left]
             del self.data[self.left]
             self.left += 1
             return elem
         def clear(self):
             self.data.clear()
             self.left = self.right = 0
         def extend(self, iterable):
             for elem in iterable:
                 self.append(elem)
         def extendleft(self, iterable):
             for elem in iterable:
                 self.appendleft(elem)
         def rotate(self, n=1):
             if self:
                 n %= len(self)
                 for i in xrange(n):
                     self.appendleft(self.pop())
         def __getitem__(self, i):
             if i < 0:
                 i += len(self)
             try:
                 return self.data[i + self.left]
             except KeyError:
                 raise IndexError
         def __setitem__(self, i, value):
             if i < 0:
                 i += len(self)
             try:
                 self.data[i + self.left] = value
             except KeyError:
                 raise IndexError
         def __delitem__(self, i):
             size = len(self)
             if not (-size <= i < size):
                 raise IndexError
             data = self.data
             if i < 0:
                 i += size
             for j in xrange(self.left + i, self.right - 1):
                 data[j] = data[j + 1]
             self.pop()
         def __len__(self):
             return self.right - self.left
         def __cmp__(self, other):
             if type(self) != type(other):
                 return cmp(type(self), type(other))
             return cmp(list(self), list(other))
         def __repr__(self, _track=[]):
             if id(self) in _track:
                 return '...'
             _track.append(id(self))
             r = 'deque(%r, maxlen=%s)' % (list(self), self.maxlen)
             _track.remove(id(self))
             return r
         def __getstate__(self):
             return (tuple(self),)
         def __setstate__(self, s):
             self.__init__(s[0])
         def __hash__(self):
             raise TypeError
         def __copy__(self):
             return self.__class__(self)
         def __deepcopy__(self, memo={}):
             from copy import deepcopy
             result = self.__class__()
             memo[id(self)] = result
             result.__init__(deepcopy(tuple(self), memo))
             return result
 #==============================================================================
 # threading.Event
 #==============================================================================
 if __py_version__ >= (2, 6):
     from threading import Event, Thread
 else:
     from threading import _Verbose, Condition, Lock, Thread, _time, \
         _allocate_lock, RLock, _sleep
     def Condition(*args, **kwargs):
         return _Condition(*args, **kwargs)
     class _Condition(_Verbose):
         def __init__(self, lock=None, verbose=None):
             _Verbose.__init__(self, verbose)
             if lock is None:
                 lock = RLock()
             self.__lock = lock
             # Export the lock's acquire() and release() methods
             self.acquire = lock.acquire
             self.release = lock.release
             # If the lock defines _release_save() and/or _acquire_restore(),
             # these override the default implementations (which just call
             # release() and acquire() on the lock).  Ditto for _is_owned().
             try:
                 self._release_save = lock._release_save
             except AttributeError:
                 pass
             try:
                 self._acquire_restore = lock._acquire_restore
             except AttributeError:
                 pass
             try:
                 self._is_owned = lock._is_owned
             except AttributeError:
                 pass
             self.__waiters = []
         def __enter__(self):
             return self.__lock.__enter__()
         def __exit__(self, *args):
             return self.__lock.__exit__(*args)
         def __repr__(self):
             return "<Condition(%s, %d)>" % (self.__lock, len(self.__waiters))
         def _release_save(self):
             self.__lock.release()           # No state to save
         def _acquire_restore(self, x):
             self.__lock.acquire()           # Ignore saved state
         def _is_owned(self):
             # Return True if lock is owned by current_thread.
             # This method is called only if __lock doesn't have _is_owned().
             if self.__lock.acquire(0):
                 self.__lock.release()
                 return False
             else:
                 return True
         def wait(self, timeout=None):
             if not self._is_owned():
                 raise RuntimeError("cannot wait on un-acquired lock")
             waiter = _allocate_lock()
             waiter.acquire()
             self.__waiters.append(waiter)
             saved_state = self._release_save()
             try:    # restore state no matter what (e.g., KeyboardInterrupt)
                 if timeout is None:
                     waiter.acquire()
                     if __debug__:
                         self._note("%s.wait(): got it", self)
                 else:
                     # Balancing act:  We can't afford a pure busy loop, so we
                     # have to sleep; but if we sleep the whole timeout time,
                     # we'll be unresponsive.  The scheme here sleeps very
                     # little at first, longer as time goes on, but never longer
                     # than 20 times per second (or the timeout time remaining).
                     endtime = _time() + timeout
                     delay = 0.0005 # 500 us -> initial delay of 1 ms
                     while True:
                         gotit = waiter.acquire(0)
                         if gotit:
                             break
                         remaining = endtime - _time()
                         if remaining <= 0:
                             break
                         delay = min(delay * 2, remaining, .05)
                         _sleep(delay)
                     if not gotit:
                         if __debug__:
                             self._note("%s.wait(%s): timed out", self, timeout)
                         try:
                             self.__waiters.remove(waiter)
                         except ValueError:
                             pass
                     else:
                         if __debug__:
                             self._note("%s.wait(%s): got it", self, timeout)
             finally:
                 self._acquire_restore(saved_state)
         def notify(self, n=1):
             if not self._is_owned():
                 raise RuntimeError("cannot notify on un-acquired lock")
             __waiters = self.__waiters
             waiters = __waiters[:n]
             if not waiters:
                 if __debug__:
                     self._note("%s.notify(): no waiters", self)
                 return
             self._note("%s.notify(): notifying %d waiter%s", self, n,
                        n != 1 and "s" or "")
             for waiter in waiters:
                 waiter.release()
                 try:
                     __waiters.remove(waiter)
                 except ValueError:
                     pass
         def notifyAll(self):
             self.notify(len(self.__waiters))
         notify_all = notifyAll
     def Event(*args, **kwargs):
         return _Event(*args, **kwargs)
     class _Event(_Verbose):
         # After Tim Peters' event class (without is_posted())
         def __init__(self, verbose=None):
             _Verbose.__init__(self, verbose)
             self.__cond = Condition(Lock())
             self.__flag = False
         def isSet(self):
             return self.__flag
         is_set = isSet
         def set(self):
             self.__cond.acquire()
             try:
                 self.__flag = True
                 self.__cond.notify_all()
             finally:
                 self.__cond.release()
         def clear(self):
             self.__cond.acquire()
             try:
                 self.__flag = False
             finally:
                 self.__cond.release()
         def wait(self, timeout=None):
             self.__cond.acquire()
             try:
                 if not self.__flag:
                     self.__cond.wait(timeout)
             finally:
                 self.__cond.release()

rhodecode/lib/middleware/pygrack.py

➞

Show inline comments

 import os
 import socket
 import logging
 import subprocess
 import traceback
 from webob import Request, Response, exc
 import rhodecode
 from rhodecode.lib import subprocessio
+from rhodecode.lib.vcs import subprocessio
 log = logging.getLogger(__name__)
 class FileWrapper(object):
     def __init__(self, fd, content_length):
         self.fd = fd
         self.content_length = content_length
         self.remain = content_length
     def read(self, size):
         if size <= self.remain:
             try:
                 data = self.fd.read(size)
             except socket.error:
                 raise IOError(self)
             self.remain -= size
         elif self.remain:
             data = self.fd.read(self.remain)
             self.remain = 0
         else:
             data = None
         return data
     def __repr__(self):
         return '<FileWrapper %s len: %s, read: %s>' % (
             self.fd, self.content_length, self.content_length - self.remain
+        )
 class GitRepository(object):
     git_folder_signature = set(['config', 'head', 'info', 'objects', 'refs'])
     commands = ['git-upload-pack', 'git-receive-pack']
     def __init__(self, repo_name, content_path, extras):
         files = set([f.lower() for f in os.listdir(content_path)])
         if  not (self.git_folder_signature.intersection(files)
                 == self.git_folder_signature):
             raise OSError('%s missing git signature' % content_path)
         self.content_path = content_path
         self.valid_accepts = ['application/x-%s-result' %
                               c for c in self.commands]
         self.repo_name = repo_name
         self.extras = extras
     def _get_fixedpath(self, path):
         """
         Small fix for repo_path
         :param path:
         :type path:
         """
         return path.split(self.repo_name, 1)[-1].strip('/')
     def inforefs(self, request, environ):
         """
         WSGI Response producer for HTTP GET Git Smart
         HTTP /info/refs request.
         """
         git_command = request.GET.get('service')
         if git_command not in self.commands:
             log.debug('command %s not allowed' % git_command)
             return exc.HTTPMethodNotAllowed()
         # note to self:
         # please, resist the urge to add '\n' to git capture and increment
         # line count by 1.
         # The code in Git client not only does NOT need '\n', but actually
         # blows up if you sprinkle "flush" (0000) as "0001\n".
         # It reads binary, per number of bytes specified.
         # if you do add '\n' as part of data, count it.
         server_advert = '# service=%s' % git_command
         packet_len = str(hex(len(server_advert) + 4)[2:].rjust(4, '0')).lower()
         _git_path = rhodecode.CONFIG.get('git_path', 'git')
         try:
             out = subprocessio.SubprocessIOChunker(
                 r'%s %s --stateless-rpc --advertise-refs "%s"' % (
                             _git_path, git_command[4:], self.content_path),
                 starting_values=[
                     packet_len + server_advert + '0000'
+                ]
+            )
         except EnvironmentError, e:
             log.error(traceback.format_exc())
             raise exc.HTTPExpectationFailed()
         resp = Response()
         resp.content_type = 'application/x-%s-advertisement' % str(git_command)
         resp.charset = None
         resp.app_iter = out
         return resp
     def backend(self, request, environ):
         """
         WSGI Response producer for HTTP POST Git Smart HTTP requests.
         Reads commands and data from HTTP POST's body.
         returns an iterator obj with contents of git command's
         response to stdout
         """
         git_command = self._get_fixedpath(request.path_info)
         if git_command not in self.commands:
             log.debug('command %s not allowed' % git_command)
             return exc.HTTPMethodNotAllowed()
         if 'CONTENT_LENGTH' in environ:
             inputstream = FileWrapper(environ['wsgi.input'],
                                       request.content_length)
         else:
             inputstream = environ['wsgi.input']
         try:
             gitenv = os.environ
             # forget all configs
             gitenv['GIT_CONFIG_NOGLOBAL'] = '1'
             opts = dict(
                 env=gitenv,
                 cwd=os.getcwd()
+            )
             cmd = r'git %s --stateless-rpc "%s"' % (git_command[4:],
                                                     self.content_path),
             log.debug('handling cmd %s' % cmd)
             out = subprocessio.SubprocessIOChunker(
                 cmd,
                 inputstream=inputstream,
                 **opts
+            )
         except EnvironmentError, e:
             log.error(traceback.format_exc())
             raise exc.HTTPExpectationFailed()
         if git_command in [u'git-receive-pack']:
             # updating refs manually after each push.
             # Needed for pre-1.7.0.4 git clients using regular HTTP mode.
             _git_path = rhodecode.CONFIG.get('git_path', 'git')
             cmd = (u'%s --git-dir "%s" '
                     'update-server-info' % (_git_path, self.content_path))
             log.debug('handling cmd %s' % cmd)
             subprocess.call(cmd, shell=True)
         resp = Response()
         resp.content_type = 'application/x-%s-result' % git_command.encode('utf8')
         resp.charset = None
         resp.app_iter = out
         return resp
     def __call__(self, environ, start_response):
         request = Request(environ)
         _path = self._get_fixedpath(request.path_info)
         if _path.startswith('info/refs'):
             app = self.inforefs
         elif [a for a in self.valid_accepts if a in request.accept]:
             app = self.backend
         try:
             resp = app(request, environ)
         except exc.HTTPException, e:
             resp = e
             log.error(traceback.format_exc())
         except Exception, e:
             log.error(traceback.format_exc())
             resp = exc.HTTPInternalServerError()
         return resp(environ, start_response)
 class GitDirectory(object):
     def __init__(self, repo_root, repo_name, extras):
         repo_location = os.path.join(repo_root, repo_name)
         if not os.path.isdir(repo_location):
             raise OSError(repo_location)
         self.content_path = repo_location
         self.repo_name = repo_name
         self.repo_location = repo_location
         self.extras = extras
     def __call__(self, environ, start_response):
         content_path = self.content_path
         try:
             app = GitRepository(self.repo_name, content_path, self.extras)
         except (AssertionError, OSError):
             content_path = os.path.join(content_path, '.git')
             if os.path.isdir(content_path):
                 app = GitRepository(self.repo_name, content_path, self.extras)
             else:
                 return exc.HTTPNotFound()(environ, start_response)
         return app(environ, start_response)
 def make_wsgi_app(repo_name, repo_root, extras):
     return GitDirectory(repo_root, repo_name, extras)

rhodecode/lib/subprocessio.py

➞

Show inline comments

deleted file

rhodecode/lib/utils.py

➞

Show inline comments

 # -*- coding: utf-8 -*-
 """
     rhodecode.lib.utils
     ~~~~~~~~~~~~~~~~~~~
     Utilities library for RhodeCode
     :created_on: Apr 18, 2010
     :author: marcink
     :copyright: (C) 2010-2012 Marcin Kuzminski <marcin@python-works.com>
     :license: GPLv3, see COPYING for more details.
 """
 # This program is free software: you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by
 # the Free Software Foundation, either version 3 of the License, or
 # (at your option) any later version.
+#
 # This program is distributed in the hope that it will be useful,
 # but WITHOUT ANY WARRANTY; without even the implied warranty of
 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 # GNU General Public License for more details.
+#
 # You should have received a copy of the GNU General Public License
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 import os
 import re
 import logging
 import datetime
 import traceback
 import paste
 import beaker
 import tarfile
 import shutil
 import decorator
 import warnings
 from os.path import abspath
 from os.path import dirname as dn, join as jn
 from paste.script.command import Command, BadCommand
 from mercurial import ui, config
 from webhelpers.text import collapse, remove_formatting, strip_tags
 from rhodecode.lib.vcs import get_backend
 from rhodecode.lib.vcs.backends.base import BaseChangeset
 from rhodecode.lib.vcs.utils.lazy import LazyProperty
 from rhodecode.lib.vcs.utils.helpers import get_scm
 from rhodecode.lib.vcs.exceptions import VCSError
 from rhodecode.lib.caching_query import FromCache
 from rhodecode.model import meta
 from rhodecode.model.db import Repository, User, RhodeCodeUi, \
     UserLog, RepoGroup, RhodeCodeSetting, CacheInvalidation, UserGroup
 from rhodecode.model.meta import Session
 from rhodecode.model.repos_group import ReposGroupModel
 from rhodecode.lib.utils2 import safe_str, safe_unicode
 from rhodecode.lib.vcs.utils.fakemod import create_module
 from rhodecode.model.users_group import UserGroupModel
 log = logging.getLogger(__name__)
 REMOVED_REPO_PAT = re.compile(r'rm__\d{8}_\d{6}_\d{6}__.*')
 def recursive_replace(str_, replace=' '):
     """
     Recursive replace of given sign to just one instance
     :param str_: given string
     :param replace: char to find and replace multiple instances
     Examples::
     >>> recursive_replace("Mighty---Mighty-Bo--sstones",'-')
     'Mighty-Mighty-Bo-sstones'
     """
     if str_.find(replace * 2) == -1:
         return str_
     else:
         str_ = str_.replace(replace * 2, replace)
         return recursive_replace(str_, replace)
 def repo_name_slug(value):
     """
     Return slug of name of repository
     This function is called on each creation/modification
     of repository to prevent bad names in repo
     """
     slug = remove_formatting(value)
     slug = strip_tags(slug)
     for c in """`?=[]\;'"<>,/~!@#$%^&*()+{}|: """:
         slug = slug.replace(c, '-')
     slug = recursive_replace(slug, '-')
     slug = collapse(slug, '-')
     return slug
 #==============================================================================
 # PERM DECORATOR HELPERS FOR EXTRACTING NAMES FOR PERM CHECKS
 #==============================================================================
 def get_repo_slug(request):
     _repo = request.environ['pylons.routes_dict'].get('repo_name')
     if _repo:
         _repo = _repo.rstrip('/')
     return _repo
 def get_repos_group_slug(request):
     _group = request.environ['pylons.routes_dict'].get('group_name')
     if _group:
         _group = _group.rstrip('/')
     return _group
 def get_user_group_slug(request):
     _group = request.environ['pylons.routes_dict'].get('id')
     try:
         _group = UserGroup.get(_group)
         if _group:
             _group = _group.users_group_name
     except Exception:
         log.debug(traceback.format_exc())
         #catch all failures here
         pass
     return _group
 def action_logger(user, action, repo, ipaddr='', sa=None, commit=False):
     """
     Action logger for various actions made by users
     :param user: user that made this action, can be a unique username string or
         object containing user_id attribute
     :param action: action to log, should be on of predefined unique actions for
         easy translations
     :param repo: string name of repository or object containing repo_id,
         that action was made on
     :param ipaddr: optional ip address from what the action was made
     :param sa: optional sqlalchemy session
     """
     if not sa:
         sa = meta.Session()
     try:
         if hasattr(user, 'user_id'):
             user_obj = User.get(user.user_id)
         elif isinstance(user, basestring):
             user_obj = User.get_by_username(user)
         else:
             raise Exception('You have to provide a user object or a username')
         if hasattr(repo, 'repo_id'):
             repo_obj = Repository.get(repo.repo_id)
             repo_name = repo_obj.repo_name
         elif  isinstance(repo, basestring):
             repo_name = repo.lstrip('/')
             repo_obj = Repository.get_by_repo_name(repo_name)
         else:
             repo_obj = None
             repo_name = ''
         user_log = UserLog()
         user_log.user_id = user_obj.user_id
         user_log.username = user_obj.username
         user_log.action = safe_unicode(action)
         user_log.repository = repo_obj
         user_log.repository_name = repo_name
         user_log.action_date = datetime.datetime.now()
         user_log.user_ip = ipaddr
         sa.add(user_log)
         log.info('Logging action:%s on %s by user:%s ip:%s' %
                  (action, safe_unicode(repo), user_obj, ipaddr))
         if commit:
             sa.commit()
     except Exception:
         log.error(traceback.format_exc())
         raise
 def get_filesystem_repos(path, recursive=False, skip_removed_repos=True):
     """
     Scans given path for repos and return (name,(type,path)) tuple
     :param path: path to scan for repositories
     :param recursive: recursive search and return names with subdirs in front
     """
     # remove ending slash for better results
     path = path.rstrip(os.sep)
     log.debug('now scanning in %s location recursive:%s...' % (path, recursive))
     def _get_repos(p):
         if not os.access(p, os.W_OK):
             log.warn('ignoring repo path without write access: %s', p)
             return
         for dirpath in os.listdir(p):
             if os.path.isfile(os.path.join(p, dirpath)):
                 continue
             cur_path = os.path.join(p, dirpath)
             # skip removed repos
             if skip_removed_repos and REMOVED_REPO_PAT.match(dirpath):
                 continue
             #skip .<somethin> dirs
             if dirpath.startswith('.'):
                 continue
             try:
                 scm_info = get_scm(cur_path)
                 yield scm_info[1].split(path, 1)[-1].lstrip(os.sep), scm_info
             except VCSError:
                 if not recursive:
                     continue
                 #check if this dir containts other repos for recursive scan
                 rec_path = os.path.join(p, dirpath)
                 if os.path.isdir(rec_path):
                     for inner_scm in _get_repos(rec_path):
                         yield inner_scm
     return _get_repos(path)
 def is_valid_repo(repo_name, base_path, scm=None):
     """
     Returns True if given path is a valid repository False otherwise.
     If scm param is given also compare if given scm is the same as expected
     from scm parameter
     :param repo_name:
     :param base_path:
     :param scm:
     :return True: if given path is a valid repository
     """
     full_path = os.path.join(safe_str(base_path), safe_str(repo_name))
     try:
         scm_ = get_scm(full_path)
         if scm:
             return scm_[0] == scm
         return True
     except VCSError:
         return False
 def is_valid_repos_group(repos_group_name, base_path, skip_path_check=False):
     """
     Returns True if given path is a repository group False otherwise
     :param repo_name:
     :param base_path:
     """
     full_path = os.path.join(safe_str(base_path), safe_str(repos_group_name))
     # check if it's not a repo
     if is_valid_repo(repos_group_name, base_path):
         return False
     try:
         # we need to check bare git repos at higher level
         # since we might match branches/hooks/info/objects or possible
         # other things inside bare git repo
         get_scm(os.path.dirname(full_path))
         return False
     except VCSError:
         pass
     # check if it's a valid path
     if skip_path_check or os.path.isdir(full_path):
         return True
     return False
 def ask_ok(prompt, retries=4, complaint='Yes or no please!'):
     while True:
         ok = raw_input(prompt)
         if ok in ('y', 'ye', 'yes'):
             return True
         if ok in ('n', 'no', 'nop', 'nope'):
             return False
         retries = retries - 1
         if retries < 0:
             raise IOError
         print complaint
 #propagated from mercurial documentation
 ui_sections = ['alias', 'auth',
                 'decode/encode', 'defaults',
                 'diff', 'email',
                 'extensions', 'format',
                 'merge-patterns', 'merge-tools',
                 'hooks', 'http_proxy',
                 'smtp', 'patch',
                 'paths', 'profiling',
                 'server', 'trusted',
                 'ui', 'web', ]
 def make_ui(read_from='file', path=None, checkpaths=True, clear_session=True):
     """
     A function that will read python rc files or database
     and make an mercurial ui object from read options
     :param path: path to mercurial config file
     :param checkpaths: check the path
     :param read_from: read from 'file' or 'db'
     """
     baseui = ui.ui()
     # clean the baseui object
     baseui._ocfg = config.config()
     baseui._ucfg = config.config()
     baseui._tcfg = config.config()
     if read_from == 'file':
         if not os.path.isfile(path):
             log.debug('hgrc file is not present at %s, skipping...' % path)
             return False
         log.debug('reading hgrc from %s' % path)
         cfg = config.config()
         cfg.read(path)
         for section in ui_sections:
             for k, v in cfg.items(section):
                 log.debug('settings ui from file: [%s] %s=%s' % (section, k, v))
                 baseui.setconfig(safe_str(section), safe_str(k), safe_str(v))
     elif read_from == 'db':
         sa = meta.Session()
         ret = sa.query(RhodeCodeUi)\
             .options(FromCache("sql_cache_short", "get_hg_ui_settings"))\
             .all()
         hg_ui = ret
         for ui_ in hg_ui:
             if ui_.ui_active:
                 log.debug('settings ui from db: [%s] %s=%s', ui_.ui_section,
                           ui_.ui_key, ui_.ui_value)
                 baseui.setconfig(safe_str(ui_.ui_section), safe_str(ui_.ui_key),
                                  safe_str(ui_.ui_value))
             if ui_.ui_key == 'push_ssl':
                 # force set push_ssl requirement to False, rhodecode
                 # handles that
                 baseui.setconfig(safe_str(ui_.ui_section), safe_str(ui_.ui_key),
                                  False)
         if clear_session:
             meta.Session.remove()
     return baseui
 def set_rhodecode_config(config):
     """
     Updates pylons config with new settings from database
     :param config:
     """
     hgsettings = RhodeCodeSetting.get_app_settings()
     for k, v in hgsettings.items():
         config[k] = v
 def set_vcs_config(config):
     """
     Patch VCS config with some RhodeCode specific stuff
     :param config: rhodecode.CONFIG
     """
     import rhodecode
     from rhodecode.lib.vcs import conf
     from rhodecode.lib.utils2 import aslist
     conf.settings.BACKENDS = {
         'hg': 'rhodecode.lib.vcs.backends.hg.MercurialRepository',
         'git': 'rhodecode.lib.vcs.backends.git.GitRepository',
+    }
     conf.settings.GIT_EXECUTABLE_PATH = config.get('git_path', 'git')
     conf.settings.GIT_REV_FILTER = config.get('git_rev_filter', '--all').strip()
     conf.settings.DEFAULT_ENCODINGS = aslist(config.get('default_encoding',
                                                         'utf8'), sep=',')
 def map_groups(path):
     """
     Given a full path to a repository, create all nested groups that this
     repo is inside. This function creates parent-child relationships between
     groups and creates default perms for all new groups.
     :param paths: full path to repository
     """
     sa = meta.Session()
     groups = path.split(Repository.url_sep())
     parent = None
     group = None
     # last element is repo in nested groups structure
     groups = groups[:-1]
     rgm = ReposGroupModel(sa)
     owner = User.get_first_admin()
     for lvl, group_name in enumerate(groups):
         group_name = '/'.join(groups[:lvl] + [group_name])
         group = RepoGroup.get_by_group_name(group_name)
         desc = '%s group' % group_name
         # skip folders that are now removed repos
         if REMOVED_REPO_PAT.match(group_name):
             break
         if group is None:
             log.debug('creating group level: %s group_name: %s'
                       % (lvl, group_name))
             group = RepoGroup(group_name, parent)
             group.group_description = desc
             group.user = owner
             sa.add(group)
             perm_obj = rgm._create_default_perms(group)
             sa.add(perm_obj)
             sa.flush()
         parent = group
     return group
 def repo2db_mapper(initial_repo_list, remove_obsolete=False,
                    install_git_hook=False):
     """
     maps all repos given in initial_repo_list, non existing repositories
     are created, if remove_obsolete is True it also check for db entries
     that are not in initial_repo_list and removes them.
     :param initial_repo_list: list of repositories found by scanning methods
     :param remove_obsolete: check for obsolete entries in database
     :param install_git_hook: if this is True, also check and install githook
         for a repo if missing
     """
     from rhodecode.model.repo import RepoModel
     from rhodecode.model.scm import ScmModel
     sa = meta.Session()
     rm = RepoModel()
     user = User.get_first_admin()
     added = []
     ##creation defaults
     defs = RhodeCodeSetting.get_default_repo_settings(strip_prefix=True)
     enable_statistics = defs.get('repo_enable_statistics')
     enable_locking = defs.get('repo_enable_locking')
     enable_downloads = defs.get('repo_enable_downloads')
     private = defs.get('repo_private')
     for name, repo in initial_repo_list.items():
         group = map_groups(name)
         db_repo = rm.get_by_repo_name(name)
         # found repo that is on filesystem not in RhodeCode database
         if not db_repo:
             log.info('repository %s not found, creating now' % name)
             added.append(name)
             desc = (repo.description
                     if repo.description != 'unknown'
                     else '%s repository' % name)
             new_repo = rm.create_repo(
                 repo_name=name,
                 repo_type=repo.alias,
                 description=desc,
                 repos_group=getattr(group, 'group_id', None),
                 owner=user,
                 just_db=True,
                 enable_locking=enable_locking,
                 enable_downloads=enable_downloads,
                 enable_statistics=enable_statistics,
                 private=private
+            )
             # we added that repo just now, and make sure it has githook
             # installed
             if new_repo.repo_type == 'git':
                 ScmModel().install_git_hook(new_repo.scm_instance)
             new_repo.update_changeset_cache()
         elif install_git_hook:
             if db_repo.repo_type == 'git':
                 ScmModel().install_git_hook(db_repo.scm_instance)
     sa.commit()
     removed = []
     if remove_obsolete:
         # remove from database those repositories that are not in the filesystem
         for repo in sa.query(Repository).all():
             if repo.repo_name not in initial_repo_list.keys():
                 log.debug("Removing non-existing repository found in db `%s`" %
                           repo.repo_name)
                 try:
                     removed.append(repo.repo_name)
                     RepoModel(sa).delete(repo, forks='detach', fs_remove=False)
                     sa.commit()
                 except Exception:
                     #don't hold further removals on error
                     log.error(traceback.format_exc())
                     sa.rollback()
     return added, removed
 # set cache regions for beaker so celery can utilise it
 def add_cache(settings):
     cache_settings = {'regions': None}
     for key in settings.keys():
         for prefix in ['beaker.cache.', 'cache.']:
             if key.startswith(prefix):
                 name = key.split(prefix)[1].strip()
                 cache_settings[name] = settings[key].strip()
     if cache_settings['regions']:
         for region in cache_settings['regions'].split(','):
             region = region.strip()
             region_settings = {}
             for key, value in cache_settings.items():
                 if key.startswith(region):
                     region_settings[key.split('.')[1]] = value
             region_settings['expire'] = int(region_settings.get('expire',
 ))
             region_settings.setdefault('lock_dir',
                                        cache_settings.get('lock_dir'))
             region_settings.setdefault('data_dir',
                                        cache_settings.get('data_dir'))
             if 'type' not in region_settings:
                 region_settings['type'] = cache_settings.get('type',
                                                              'memory')
             beaker.cache.cache_regions[region] = region_settings
 def load_rcextensions(root_path):
     import rhodecode
     from rhodecode.config import conf
     path = os.path.join(root_path, 'rcextensions', '__init__.py')
     if os.path.isfile(path):
         rcext = create_module('rc', path)
         EXT = rhodecode.EXTENSIONS = rcext
         log.debug('Found rcextensions now loading %s...' % rcext)
         # Additional mappings that are not present in the pygments lexers
         conf.LANGUAGES_EXTENSIONS_MAP.update(getattr(EXT, 'EXTRA_MAPPINGS', {}))
         #OVERRIDE OUR EXTENSIONS FROM RC-EXTENSIONS (if present)
         if getattr(EXT, 'INDEX_EXTENSIONS', []) != []:
             log.debug('settings custom INDEX_EXTENSIONS')
             conf.INDEX_EXTENSIONS = getattr(EXT, 'INDEX_EXTENSIONS', [])
         #ADDITIONAL MAPPINGS
         log.debug('adding extra into INDEX_EXTENSIONS')
         conf.INDEX_EXTENSIONS.extend(getattr(EXT, 'EXTRA_INDEX_EXTENSIONS', []))
         # auto check if the module is not missing any data, set to default if is
         # this will help autoupdate new feature of rcext module
         from rhodecode.config import rcextensions
         for k in dir(rcextensions):
             if not k.startswith('_') and not hasattr(EXT, k):
                 setattr(EXT, k, getattr(rcextensions, k))
 def get_custom_lexer(extension):
     """
     returns a custom lexer if it's defined in rcextensions module, or None
     if there's no custom lexer defined
     """
     import rhodecode
     from pygments import lexers
     #check if we didn't define this extension as other lexer
     if rhodecode.EXTENSIONS and extension in rhodecode.EXTENSIONS.EXTRA_LEXERS:
         _lexer_name = rhodecode.EXTENSIONS.EXTRA_LEXERS[extension]
         return lexers.get_lexer_by_name(_lexer_name)
 #==============================================================================
 # TEST FUNCTIONS AND CREATORS
 #==============================================================================
 def create_test_index(repo_location, config, full_index):
     """
     Makes default test index
     :param config: test config
     :param full_index:
     """
     from rhodecode.lib.indexers.daemon import WhooshIndexingDaemon
     from rhodecode.lib.pidlock import DaemonLock, LockHeld
     repo_location = repo_location
     index_location = os.path.join(config['app_conf']['index_dir'])
     if not os.path.exists(index_location):
         os.makedirs(index_location)
     try:
         l = DaemonLock(file_=jn(dn(index_location), 'make_index.lock'))
         WhooshIndexingDaemon(index_location=index_location,
                              repo_location=repo_location)\
             .run(full_index=full_index)
         l.release()
     except LockHeld:
         pass
 def create_test_env(repos_test_path, config):
     """
     Makes a fresh database and
     install test repository into tmp dir
     """
     from rhodecode.lib.db_manage import DbManage
     from rhodecode.tests import HG_REPO, GIT_REPO, TESTS_TMP_PATH
     # PART ONE create db
     dbconf = config['sqlalchemy.db1.url']
     log.debug('making test db %s' % dbconf)
     # create test dir if it doesn't exist
     if not os.path.isdir(repos_test_path):
         log.debug('Creating testdir %s' % repos_test_path)
         os.makedirs(repos_test_path)
     dbmanage = DbManage(log_sql=True, dbconf=dbconf, root=config['here'],
                         tests=True)
     dbmanage.create_tables(override=True)
     dbmanage.create_settings(dbmanage.config_prompt(repos_test_path))
     dbmanage.create_default_user()
     dbmanage.admin_prompt()
     dbmanage.create_permissions()
     dbmanage.populate_default_permissions()
     Session().commit()
     # PART TWO make test repo
     log.debug('making test vcs repositories')
     idx_path = config['app_conf']['index_dir']
     data_path = config['app_conf']['cache_dir']
     #clean index and data
     if idx_path and os.path.exists(idx_path):
         log.debug('remove %s' % idx_path)
         shutil.rmtree(idx_path)
     if data_path and os.path.exists(data_path):
         log.debug('remove %s' % data_path)
         shutil.rmtree(data_path)
     #CREATE DEFAULT TEST REPOS
     cur_dir = dn(dn(abspath(__file__)))
     tar = tarfile.open(jn(cur_dir, 'tests', "vcs_test_hg.tar.gz"))
     tar.extractall(jn(TESTS_TMP_PATH, HG_REPO))
     tar.close()
     cur_dir = dn(dn(abspath(__file__)))
     tar = tarfile.open(jn(cur_dir, 'tests', "vcs_test_git.tar.gz"))
     tar.extractall(jn(TESTS_TMP_PATH, GIT_REPO))
     tar.close()
     #LOAD VCS test stuff
     from rhodecode.tests.vcs import setup_package
     setup_package()
 #==============================================================================
 # PASTER COMMANDS
 #==============================================================================
 class BasePasterCommand(Command):
     """
     Abstract Base Class for paster commands.
     The celery commands are somewhat aggressive about loading
     celery.conf, and since our module sets the `CELERY_LOADER`
     environment variable to our loader, we have to bootstrap a bit and
     make sure we've had a chance to load the pylons config off of the
     command line, otherwise everything fails.
     """
     min_args = 1
     min_args_error = "Please provide a paster config file as an argument."
     takes_config_file = 1
     requires_config_file = True
     def notify_msg(self, msg, log=False):
         """Make a notification to user, additionally if logger is passed
         it logs this action using given logger
         :param msg: message that will be printed to user
         :param log: logging instance, to use to additionally log this message
         """
         if log and isinstance(log, logging):
             log(msg)
     def run(self, args):
         """
         Overrides Command.run
         Checks for a config file argument and loads it.
         """
         if len(args) < self.min_args:
             raise BadCommand(
                 self.min_args_error % {'min_args': self.min_args,
                                        'actual_args': len(args)})
         # Decrement because we're going to lob off the first argument.
         # @@ This is hacky
         self.min_args -= 1
         self.bootstrap_config(args[0])
         self.update_parser()
         return super(BasePasterCommand, self).run(args[1:])
     def update_parser(self):
         """
         Abstract method.  Allows for the class's parser to be updated
         before the superclass's `run` method is called.  Necessary to
         allow options/arguments to be passed through to the underlying
         celery command.
         """
         raise NotImplementedError("Abstract Method.")
     def bootstrap_config(self, conf):
         """
         Loads the pylons configuration.
         """
         from pylons import config as pylonsconfig
         self.path_to_ini_file = os.path.realpath(conf)
         conf = paste.deploy.appconfig('config:' + self.path_to_ini_file)
         pylonsconfig.init_app(conf.global_conf, conf.local_conf)
     def _init_session(self):
         """
         Inits SqlAlchemy Session
         """
         logging.config.fileConfig(self.path_to_ini_file)
         from pylons import config
         from rhodecode.model import init_model
         from rhodecode.lib.utils2 import engine_from_config
         #get to remove repos !!
         add_cache(config)
         engine = engine_from_config(config, 'sqlalchemy.db1.')
         init_model(engine)
 def check_git_version():
     """
     Checks what version of git is installed in system, and issues a warning
     if it's too old for RhodeCode to properly work.
     """
     from rhodecode import BACKENDS
     from rhodecode.lib.vcs.backends.git.repository import GitRepository
     from rhodecode.lib.vcs.conf import settings
     from distutils.version import StrictVersion
     stdout, stderr = GitRepository._run_git_command('--version', _bare=True,
                                                     _safe=True)
     ver = (stdout.split(' ')[-1] or '').strip() or '0.0.0'
     if len(ver.split('.')) > 3:
         #StrictVersion needs to be only 3 element type
         ver = '.'.join(ver.split('.')[:3])
     try:
         _ver = StrictVersion(ver)
     except Exception:
         _ver = StrictVersion('0.0.0')
         stderr = traceback.format_exc()
     req_ver = '1.7.4'
     to_old_git = False
     if  _ver < StrictVersion(req_ver):
         to_old_git = True
     if 'git' in BACKENDS:
         log.debug('GIT version detected: %s' % stdout)
         log.debug('GIT executable: "%s" version detected: %s'
                   % (settings.GIT_EXECUTABLE_PATH, stdout))
         if stderr:
             log.warning('Unable to detect git version, org error was: %r' % stderr)
         elif to_old_git:
             log.warning('RhodeCode detected git version %s, which is too old '
                         'for the system to function properly. Make sure '
                         'its version is at least %s' % (ver, req_ver))
     return _ver
 @decorator.decorator
 def jsonify(func, *args, **kwargs):
     """Action decorator that formats output for JSON
     Given a function that will return content, this decorator will turn
     the result into JSON, with a content-type of 'application/json' and
     output it.
     """
     from pylons.decorators.util import get_pylons
     from rhodecode.lib.compat import json
     pylons = get_pylons(args)
     pylons.response.headers['Content-Type'] = 'application/json; charset=utf-8'
     data = func(*args, **kwargs)
     if isinstance(data, (list, tuple)):
         msg = "JSON responses with Array envelopes are susceptible to " \
               "cross-site data leak attacks, see " \
               "http://wiki.pylonshq.com/display/pylonsfaq/Warnings"
         warnings.warn(msg, Warning, 2)
         log.warning(msg)
     log.debug("Returning JSON wrapped action output")
     return json.dumps(data, encoding='utf-8')

rhodecode/lib/vcs/__init__.py

➞

Show inline comments

 # -*- 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, 3, 0, 'dev')
+VERSION = (0, 4, 0, 'dev')
 __version__ = '.'.join((str(each) for each in VERSION[:4]))
 __all__ = [
     'get_version', 'get_repo', 'get_backend',
     'VCSError', 'RepositoryError', 'ChangesetError']
     '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/base.py

➞

Show inline comments

 # -*- coding: utf-8 -*-
 """
     vcs.backends.base
     ~~~~~~~~~~~~~~~~~
     Base for all available scm backends
     :created_on: Apr 8, 2010
     :copyright: (c) 2010-2011 by Marcin Kuzminski, Lukasz Balcerzak.
 """
 import datetime
 from itertools import chain
 import itertools
 from rhodecode.lib.vcs.utils import author_name, author_email
 from rhodecode.lib.vcs.utils.lazy import LazyProperty
 from rhodecode.lib.vcs.utils.helpers import get_dict_for_attrs
 from rhodecode.lib.vcs.conf import settings
 from rhodecode.lib.vcs.exceptions import ChangesetError, EmptyRepositoryError, \
     NodeAlreadyAddedError, NodeAlreadyChangedError, NodeAlreadyExistsError, \
     NodeAlreadyRemovedError, NodeDoesNotExistError, NodeNotChangedError, \
     RepositoryError
 from rhodecode.lib.vcs.exceptions import (
     ChangesetError, EmptyRepositoryError, NodeAlreadyAddedError,
     NodeAlreadyChangedError, NodeAlreadyExistsError, NodeAlreadyRemovedError,
     NodeDoesNotExistError, NodeNotChangedError, RepositoryError
+)
 class BaseRepository(object):
     """
     Base Repository for final backends
     **Attributes**
         ``DEFAULT_BRANCH_NAME``
             name of default branch (i.e. "trunk" for svn, "master" for git etc.
         ``scm``
             alias of scm, i.e. *git* or *hg*
         ``repo``
             object from external api
         ``revisions``
             list of all available revisions' ids, in ascending order
         ``changesets``
             storage dict caching returned changesets
         ``path``
             absolute path to the repository
         ``branches``
             branches as list of changesets
         ``tags``
             tags as list of changesets
     """
     scm = None
     DEFAULT_BRANCH_NAME = None
     EMPTY_CHANGESET = '0' * 40
     def __init__(self, repo_path, create=False, **kwargs):
         """
         Initializes repository. Raises RepositoryError if repository could
         not be find at the given ``repo_path`` or directory at ``repo_path``
         exists and ``create`` is set to True.
         :param repo_path: local path of the repository
         :param create=False: if set to True, would try to craete repository.
         :param src_url=None: if set, should be proper url from which repository
           would be cloned; requires ``create`` parameter to be set to True -
           raises RepositoryError if src_url is set and create evaluates to
           False
         """
         raise NotImplementedError
     def __str__(self):
         return '<%s at %s>' % (self.__class__.__name__, self.path)
     def __repr__(self):
         return self.__str__()
     def __len__(self):
         return self.count()
     def __eq__(self, other):
         same_instance = isinstance(other, self.__class__)
         return same_instance and getattr(other, 'path', None) == self.path
     def __ne__(self, other):
         return not self.__eq__(other)
     @LazyProperty
     def alias(self):
         for k, v in settings.BACKENDS.items():
             if v.split('.')[-1] == str(self.__class__.__name__):
                 return k
     @LazyProperty
     def name(self):
         raise NotImplementedError
     @LazyProperty
     def owner(self):
         raise NotImplementedError
     @LazyProperty
     def description(self):
         raise NotImplementedError
     @LazyProperty
     def size(self):
         """
         Returns combined size in bytes for all repository files
         """
         size = 0
         try:
             tip = self.get_changeset()
             for topnode, dirs, files in tip.walk('/'):
                 for f in files:
                     size += tip.get_file_size(f.path)
                 for dir in dirs:
                     for f in files:
                         size += tip.get_file_size(f.path)
         except RepositoryError, e:
             pass
         return size
     def is_valid(self):
         """
         Validates repository.
         """
         raise NotImplementedError
     def get_last_change(self):
         self.get_changesets()
     #==========================================================================
     # CHANGESETS
     #==========================================================================
     def get_changeset(self, revision=None):
         """
         Returns instance of ``Changeset`` class. If ``revision`` is None, most
         recent changeset is returned.
         :raises ``EmptyRepositoryError``: if there are no revisions
         """
         raise NotImplementedError
     def __iter__(self):
         """
         Allows Repository objects to be iterated.
         *Requires* implementation of ``__getitem__`` method.
         """
         for revision in self.revisions:
             yield self.get_changeset(revision)
     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
         not inclusive This should behave just like a list, ie. end is not
         inclusive
         :param start: None or str
         :param end: None or str
         :param start_date:
         :param end_date:
         :param branch_name:
         :param reversed:
         """
         raise NotImplementedError
     def __getslice__(self, i, j):
         """
         Returns a iterator of sliced repository
         """
         for rev in self.revisions[i:j]:
             yield self.get_changeset(rev)
     def __getitem__(self, key):
         return self.get_changeset(key)
     def count(self):
         return len(self.revisions)
     def tag(self, name, user, revision=None, message=None, date=None, **opts):
         """
         Creates and returns a tag for the given ``revision``.
         :param name: name for new tag
         :param user: full username, i.e.: "Joe Doe <joe.doe@example.com>"
         :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
         """
         raise NotImplementedError
     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 <joe.doe@example.com>"
         :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
         """
         raise NotImplementedError
     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``.
         """
         raise NotImplementedError
     # ========== #
     # COMMIT API #
     # ========== #
     @LazyProperty
     def in_memory_changeset(self):
         """
         Returns ``InMemoryChangeset`` object for this repository.
         """
         raise NotImplementedError
     def add(self, filenode, **kwargs):
         """
         Commit api function that will add given ``FileNode`` into this
         repository.
         :raises ``NodeAlreadyExistsError``: if there is a file with same path
           already in repository
         :raises ``NodeAlreadyAddedError``: if given node is already marked as
           *added*
         """
         raise NotImplementedError
     def remove(self, filenode, **kwargs):
         """
         Commit api function that will remove given ``FileNode`` into this
         repository.
         :raises ``EmptyRepositoryError``: if there are no changesets yet
         :raises ``NodeDoesNotExistError``: if there is no file with given path
         """
         raise NotImplementedError
     def commit(self, message, **kwargs):
         """
         Persists current changes made on this repository and returns newly
         created changeset.
         :raises ``NothingChangedError``: if no changes has been made
         """
         raise NotImplementedError
     def get_state(self):
         """
         Returns dictionary with ``added``, ``changed`` and ``removed`` lists
         containing ``FileNode`` objects.
         """
         raise NotImplementedError
     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)
         """
         raise NotImplementedError
     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)
         """
         raise NotImplementedError
     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)
         """
         raise NotImplementedError
     # =========== #
     # WORKDIR API #
     # =========== #
     @LazyProperty
     def workdir(self):
         """
         Returns ``Workdir`` instance for this repository.
         """
         raise NotImplementedError
 class BaseChangeset(object):
     """
     Each backend should implement it's changeset representation.
     **Attributes**
         ``repository``
             repository object within which changeset exists
         ``id``
             may be ``raw_id`` or i.e. for mercurial's tip just ``tip``
         ``raw_id``
             raw changeset representation (i.e. full 40 length sha for git
             backend)
         ``short_id``
             shortened (if apply) version of ``raw_id``; it would be simple
             shortcut for ``raw_id[:12]`` for git/mercurial backends or same
             as ``raw_id`` for subversion
         ``revision``
             revision number as integer
         ``files``
             list of ``FileNode`` (``Node`` with NodeKind.FILE) objects
         ``dirs``
             list of ``DirNode`` (``Node`` with NodeKind.DIR) objects
         ``nodes``
             combined list of ``Node`` objects
         ``author``
             author of the changeset, as unicode
         ``message``
             message of the changeset, as unicode
         ``parents``
             list of parent changesets
         ``last``
             ``True`` if this is last changeset in repository, ``False``
             otherwise; trying to access this attribute while there is no
             changesets would raise ``EmptyRepositoryError``
     """
     def __str__(self):
         return '<%s at %s:%s>' % (self.__class__.__name__, self.revision,
             self.short_id)
     def __repr__(self):
         return self.__str__()
     def __unicode__(self):
         return u'%s:%s' % (self.revision, self.short_id)
     def __eq__(self, other):
         return self.raw_id == other.raw_id
     def __json__(self):
         return dict(
             short_id=self.short_id,
             raw_id=self.raw_id,
             revision=self.revision,
             message=self.message,
             date=self.date,
             author=self.author,
+        )
     @LazyProperty
     def last(self):
         if self.repository is None:
             raise ChangesetError("Cannot check if it's most recent revision")
         return self.raw_id == self.repository.revisions[-1]
     @LazyProperty
     def parents(self):
         """
         Returns list of parents changesets.
         """
         raise NotImplementedError
     @LazyProperty
     def children(self):
@@ @@ -470,559 +472,559 @@ class BaseChangeset(object): @@
         """
         raise NotImplementedError
     @LazyProperty
     def author_name(self):
         """
         Returns Author name for given commit
         """
         return author_name(self.author)
     @LazyProperty
     def author_email(self):
         """
         Returns Author email address for given commit
         """
         return author_email(self.author)
     def get_file_mode(self, path):
         """
         Returns stat mode of the file at the given ``path``.
         """
         raise NotImplementedError
     def get_file_content(self, path):
         """
         Returns content of the file at the given ``path``.
         """
         raise NotImplementedError
     def get_file_size(self, path):
         """
         Returns size of the file at the given ``path``.
         """
         raise NotImplementedError
     def get_file_changeset(self, path):
         """
         Returns last commit of the file at the given ``path``.
         """
         raise NotImplementedError
     def get_file_history(self, path):
         """
         Returns history of file as reversed list of ``Changeset`` objects for
         which file at given ``path`` has been modified.
         """
         raise NotImplementedError
     def get_nodes(self, path):
         """
         Returns combined ``DirNode`` and ``FileNode`` objects list representing
         state of changeset at the given ``path``.
         :raises ``ChangesetError``: if node at the given ``path`` is not
           instance of ``DirNode``
         """
         raise NotImplementedError
     def get_node(self, path):
         """
         Returns ``Node`` object from the given ``path``.
         :raises ``NodeDoesNotExistError``: if there is no node at the given
           ``path``
         """
         raise NotImplementedError
     def fill_archive(self, stream=None, kind='tgz', prefix=None):
         """
         Fills up given stream.
         :param stream: file like object.
         :param kind: one of following: ``zip``, ``tar``, ``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>
         """
         raise NotImplementedError
     def get_chunked_archive(self, **kwargs):
         """
         Returns iterable archive. Tiny wrapper around ``fill_archive`` method.
         :param chunk_size: extra parameter which controls size of returned
             chunks. Default:8k.
         """
         chunk_size = kwargs.pop('chunk_size', 8192)
         stream = kwargs.get('stream')
         self.fill_archive(**kwargs)
         while True:
             data = stream.read(chunk_size)
             if not data:
                 break
             yield data
     @LazyProperty
     def root(self):
         """
         Returns ``RootNode`` object for this changeset.
         """
         return self.get_node('')
     def next(self, branch=None):
         """
         Returns next changeset from current, if branch is gives it will return
         next changeset belonging to this branch
         :param branch: show changesets within the given named branch
         """
         raise NotImplementedError
     def prev(self, branch=None):
         """
         Returns previous changeset from current, if branch is gives it will
         return previous changeset belonging to this branch
         :param branch: show changesets within the given named branch
         """
         raise NotImplementedError
     @LazyProperty
     def added(self):
         """
         Returns list of added ``FileNode`` objects.
         """
         raise NotImplementedError
     @LazyProperty
     def changed(self):
         """
         Returns list of modified ``FileNode`` objects.
         """
         raise NotImplementedError
     @LazyProperty
     def removed(self):
         """
         Returns list of removed ``FileNode`` objects.
         """
         raise NotImplementedError
     @LazyProperty
     def size(self):
         """
         Returns total number of bytes from contents of all filenodes.
         """
         return sum((node.size for node in self.get_filenodes_generator()))
     def walk(self, topurl=''):
         """
         Similar to os.walk method. Insted of filesystem it walks through
         changeset starting at given ``topurl``.  Returns generator of tuples
         (topnode, dirnodes, filenodes).
         """
         topnode = self.get_node(topurl)
         yield (topnode, topnode.dirs, topnode.files)
         for dirnode in topnode.dirs:
             for tup in self.walk(dirnode.path):
                 yield tup
     def get_filenodes_generator(self):
         """
         Returns generator that yields *all* file nodes.
         """
         for topnode, dirs, files in self.walk():
             for node in files:
                 yield node
     def as_dict(self):
         """
         Returns dictionary with changeset's attributes and their values.
         """
         data = get_dict_for_attrs(self, ['id', 'raw_id', 'short_id',
             'revision', 'date', 'message'])
         data['author'] = {'name': self.author_name, 'email': self.author_email}
         data['added'] = [node.path for node in self.added]
         data['changed'] = [node.path for node in self.changed]
         data['removed'] = [node.path for node in self.removed]
         return data
 class BaseWorkdir(object):
     """
     Working directory representation of single repository.
     :attribute: repository: repository object of working directory
     """
     def __init__(self, repository):
         self.repository = repository
     def get_branch(self):
         """
         Returns name of current branch.
         """
         raise NotImplementedError
     def get_changeset(self):
         """
         Returns current changeset.
         """
         raise NotImplementedError
     def get_added(self):
         """
         Returns list of ``FileNode`` objects marked as *new* in working
         directory.
         """
         raise NotImplementedError
     def get_changed(self):
         """
         Returns list of ``FileNode`` objects *changed* in working directory.
         """
         raise NotImplementedError
     def get_removed(self):
         """
         Returns list of ``RemovedFileNode`` objects marked as *removed* in
         working directory.
         """
         raise NotImplementedError
     def get_untracked(self):
         """
         Returns list of ``FileNode`` objects which are present within working
         directory however are not tracked by repository.
         """
         raise NotImplementedError
     def get_status(self):
         """
         Returns dict with ``added``, ``changed``, ``removed`` and ``untracked``
         lists.
         """
         raise NotImplementedError
     def commit(self, message, **kwargs):
         """
         Commits local (from working directory) changes and returns newly
         created
         ``Changeset``. Updates repository's ``revisions`` list.
         :raises ``CommitError``: if any error occurs while committing
         """
         raise NotImplementedError
     def update(self, revision=None):
         """
         Fetches content of the given revision and populates it within working
         directory.
         """
         raise NotImplementedError
     def checkout_branch(self, branch=None):
         """
         Checks out ``branch`` or the backend's default branch.
         Raises ``BranchDoesNotExistError`` if the branch does not exist.
         """
         raise NotImplementedError
 class BaseInMemoryChangeset(object):
     """
     Represents differences between repository's state (most recent head) and
     changes made *in place*.
     **Attributes**
         ``repository``
             repository object for this in-memory-changeset
         ``added``
             list of ``FileNode`` objects marked as *added*
         ``changed``
             list of ``FileNode`` objects marked as *changed*
         ``removed``
             list of ``FileNode`` or ``RemovedFileNode`` objects marked to be
             *removed*
         ``parents``
             list of ``Changeset`` representing parents of in-memory changeset.
             Should always be 2-element sequence.
     """
     def __init__(self, repository):
         self.repository = repository
         self.added = []
         self.changed = []
         self.removed = []
         self.parents = []
     def add(self, *filenodes):
         """
         Marks given ``FileNode`` objects as *to be committed*.
         :raises ``NodeAlreadyExistsError``: if node with same path exists at
           latest changeset
         :raises ``NodeAlreadyAddedError``: if node with same path is already
           marked as *added*
         """
         # Check if not already marked as *added* first
         for node in filenodes:
             if node.path in (n.path for n in self.added):
                 raise NodeAlreadyAddedError("Such FileNode %s is already "
                     "marked for addition" % node.path)
         for node in filenodes:
             self.added.append(node)
     def change(self, *filenodes):
         """
         Marks given ``FileNode`` objects to be *changed* in next commit.
         :raises ``EmptyRepositoryError``: if there are no changesets yet
         :raises ``NodeAlreadyExistsError``: if node with same path is already
           marked to be *changed*
         :raises ``NodeAlreadyRemovedError``: if node with same path is already
           marked to be *removed*
         :raises ``NodeDoesNotExistError``: if node doesn't exist in latest
           changeset
         :raises ``NodeNotChangedError``: if node hasn't really be changed
         """
         for node in filenodes:
             if node.path in (n.path for n in self.removed):
                 raise NodeAlreadyRemovedError("Node at %s is already marked "
                     "as removed" % node.path)
         try:
             self.repository.get_changeset()
         except EmptyRepositoryError:
             raise EmptyRepositoryError("Nothing to change - try to *add* new "
                 "nodes rather than changing them")
         for node in filenodes:
             if node.path in (n.path for n in self.changed):
                 raise NodeAlreadyChangedError("Node at '%s' is already "
                     "marked as changed" % node.path)
             self.changed.append(node)
     def remove(self, *filenodes):
         """
         Marks given ``FileNode`` (or ``RemovedFileNode``) objects to be
         *removed* in next commit.
         :raises ``NodeAlreadyRemovedError``: if node has been already marked to
           be *removed*
         :raises ``NodeAlreadyChangedError``: if node has been already marked to
           be *changed*
         """
         for node in filenodes:
             if node.path in (n.path for n in self.removed):
                 raise NodeAlreadyRemovedError("Node is already marked to "
                     "for removal at %s" % node.path)
             if node.path in (n.path for n in self.changed):
                 raise NodeAlreadyChangedError("Node is already marked to "
                     "be changed at %s" % node.path)
             # We only mark node as *removed* - real removal is done by
             # commit method
             self.removed.append(node)
     def reset(self):
         """
         Resets this instance to initial state (cleans ``added``, ``changed``
         and ``removed`` lists).
         """
         self.added = []
         self.changed = []
         self.removed = []
         self.parents = []
     def get_ipaths(self):
         """
         Returns generator of paths from nodes marked as added, changed or
         removed.
         """
         for node in chain(self.added, self.changed, self.removed):
+        for node in itertools.chain(self.added, self.changed, self.removed):
             yield node.path
     def get_paths(self):
         """
         Returns list of paths from nodes marked as added, changed or removed.
         """
         return list(self.get_ipaths())
     def check_integrity(self, parents=None):
         """
         Checks in-memory changeset's integrity. Also, sets parents if not
         already set.
         :raises CommitError: if any error occurs (i.e.
           ``NodeDoesNotExistError``).
         """
         if not self.parents:
             parents = parents or []
             if len(parents) == 0:
                 try:
                     parents = [self.repository.get_changeset(), None]
                 except EmptyRepositoryError:
                     parents = [None, None]
             elif len(parents) == 1:
                 parents += [None]
             self.parents = parents
         # Local parents, only if not None
         parents = [p for p in self.parents if p]
         # Check nodes marked as added
         for p in parents:
             for node in self.added:
                 try:
                     p.get_node(node.path)
                 except NodeDoesNotExistError:
                     pass
                 else:
                     raise NodeAlreadyExistsError("Node at %s already exists "
                         "at %s" % (node.path, p))
         # Check nodes marked as changed
         missing = set(self.changed)
         not_changed = set(self.changed)
         if self.changed and not parents:
             raise NodeDoesNotExistError(str(self.changed[0].path))
         for p in parents:
             for node in self.changed:
                 try:
                     old = p.get_node(node.path)
                     missing.remove(node)
                     if old.content != node.content:
                         not_changed.remove(node)
                 except NodeDoesNotExistError:
                     pass
         if self.changed and missing:
             raise NodeDoesNotExistError("Node at %s is missing "
                 "(parents: %s)" % (node.path, parents))
         if self.changed and not_changed:
             raise NodeNotChangedError("Node at %s wasn't actually changed "
                 "since parents' changesets: %s" % (not_changed.pop().path,
                     parents)
+            )
         # Check nodes marked as removed
         if self.removed and not parents:
             raise NodeDoesNotExistError("Cannot remove node at %s as there "
                 "were no parents specified" % self.removed[0].path)
         really_removed = set()
         for p in parents:
             for node in self.removed:
                 try:
                     p.get_node(node.path)
                     really_removed.add(node)
                 except ChangesetError:
                     pass
         not_removed = set(self.removed) - really_removed
         if not_removed:
             raise NodeDoesNotExistError("Cannot remove node at %s from "
                 "following parents: %s" % (not_removed[0], parents))
     def commit(self, message, author, parents=None, branch=None, date=None,
             **kwargs):
         """
         Performs in-memory commit (doesn't check workdir in any way) and
         returns newly created ``Changeset``. Updates repository's
         ``revisions``.
         .. note::
             While overriding this method each backend's should call
             ``self.check_integrity(parents)`` in the first place.
         :param message: message of the commit
         :param author: full username, i.e. "Joe Doe <joe.doe@example.com>"
         :param parents: single parent or sequence of parents from which commit
           would be derieved
         :param date: ``datetime.datetime`` instance. Defaults to
           ``datetime.datetime.now()``.
         :param branch: branch name, as string. If none given, default backend's
           branch would be used.
         :raises ``CommitError``: if any error occurs while committing
         """
         raise NotImplementedError
 class EmptyChangeset(BaseChangeset):
     """
     An dummy empty changeset. It's possible to pass hash when creating
     an EmptyChangeset
     """
     def __init__(self, cs='0' * 40, repo=None, requested_revision=None,
                  alias=None, revision=-1, message='', author='', date=None):
         self._empty_cs = cs
         self.revision = revision
         self.message = message
         self.author = author
         self.date = date or datetime.datetime.fromtimestamp(0)
         self.repository = repo
         self.requested_revision = requested_revision
         self.alias = alias
     @LazyProperty
     def raw_id(self):
         """
         Returns raw string identifying this changeset, useful for web
         representation.
         """
         return self._empty_cs
     @LazyProperty
     def branch(self):
         from rhodecode.lib.vcs.backends import get_backend
         return get_backend(self.alias).DEFAULT_BRANCH_NAME
     @LazyProperty
     def short_id(self):
         return self.raw_id[:12]
     def get_file_changeset(self, path):
         return self
     def get_file_content(self, path):
         return u''
     def get_file_size(self, path):
         return 0
 class CollectionGenerator(object):
     def __init__(self, repo, revs):
         self.repo = repo
         self.revs = revs
     def __len__(self):
         return len(self.revs)
     def __iter__(self):
         for rev in self.revs:
             yield self.repo.get_changeset(rev)
     def __getslice__(self, i, j):
         """
         Returns a iterator of sliced repository
         """
         sliced_revs = self.revs[i:j]
         return CollectionGenerator(self.repo, sliced_revs)
     def __repr__(self):
         return 'CollectionGenerator<%s>' % (len(self))

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

➞

Show inline comments

 import re
 from itertools import chain
 from dulwich import objects
 from subprocess import Popen, PIPE
-import rhodecode
 from rhodecode.lib.vcs.conf import settings
 from rhodecode.lib.vcs.exceptions import RepositoryError
 from rhodecode.lib.vcs.exceptions import ChangesetError
 from rhodecode.lib.vcs.exceptions import NodeDoesNotExistError
 from rhodecode.lib.vcs.exceptions import VCSError
 from rhodecode.lib.vcs.exceptions import ChangesetDoesNotExistError
 from rhodecode.lib.vcs.exceptions import ImproperArchiveTypeError
 from rhodecode.lib.vcs.backends.base import BaseChangeset, EmptyChangeset
 from rhodecode.lib.vcs.nodes import FileNode, DirNode, NodeKind, RootNode, \
     RemovedFileNode, SubModuleNode, ChangedFileNodesGenerator,\
     AddedFileNodesGenerator, RemovedFileNodesGenerator
 from rhodecode.lib.vcs.utils import safe_unicode
 from rhodecode.lib.vcs.utils import date_fromtimestamp
 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
 from rhodecode.lib.utils2 import safe_int, safe_str
 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)
             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 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 = rhodecode.CONFIG.get('git_rev_filter',
                                               '--all').strip()
         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 = rhodecode.CONFIG.get('git_path', 'git')
+        _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
         """
         a, m, d = self._changes_cache
         return list(a.union(m).union(d))
         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*
         """
-        a, m, d = self._changes_cache
+        added, modified, deleted = self._changes_cache
         return sorted({
             'added': list(a),
             'modified': list(m),
             'deleted': list(d)}[status]
             '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/inmemory.py

➞

Show inline comments

 import time
 import datetime
 import posixpath
 from dulwich import objects
 from dulwich.repo import Repo
 from rhodecode.lib.vcs.backends.base import BaseInMemoryChangeset
 from rhodecode.lib.vcs.exceptions import RepositoryError
 from rhodecode.lib.vcs.utils import safe_str
 class GitInMemoryChangeset(BaseInMemoryChangeset):
     def commit(self, message, author, parents=None, branch=None, date=None,
                **kwargs):
         """
         Performs in-memory commit (doesn't check workdir in any way) and
         returns newly created ``Changeset``. Updates repository's
         ``revisions``.
         :param message: message of the commit
         :param author: full username, i.e. "Joe Doe <joe.doe@example.com>"
         :param parents: single parent or sequence of parents from which commit
           would be derieved
         :param date: ``datetime.datetime`` instance. Defaults to
           ``datetime.datetime.now()``.
         :param branch: branch name, as string. If none given, default backend's
           branch would be used.
         :raises ``CommitError``: if any error occurs while committing
         """
         self.check_integrity(parents)
         from .repository import GitRepository
         if branch is None:
             branch = GitRepository.DEFAULT_BRANCH_NAME
         repo = self.repository._repo
         object_store = repo.object_store
         ENCODING = "UTF-8"
         DIRMOD = 040000
         # Create tree and populates it with blobs
         commit_tree = self.parents[0] and repo[self.parents[0]._commit.tree] or\
             objects.Tree()
         for node in self.added + self.changed:
             # Compute subdirs if needed
             dirpath, nodename = posixpath.split(node.path)
             dirnames = dirpath and dirpath.split('/') or []
             parent = commit_tree
             ancestors = [('', parent)]
             # Tries to dig for the deepest existing tree
             while dirnames:
                 curdir = dirnames.pop(0)
                 try:
                     dir_id = parent[curdir][1]
                 except KeyError:
                     # put curdir back into dirnames and stops
                     dirnames.insert(0, curdir)
                     break
                 else:
                     # If found, updates parent
                     parent = self.repository._repo[dir_id]
                     ancestors.append((curdir, parent))
             # Now parent is deepest existing tree and we need to create subtrees
             # for dirnames (in reverse order) [this only applies for nodes from added]
             new_trees = []
             if not node.is_binary:
                 content = node.content.encode(ENCODING)
             else:
                 content = node.content
             blob = objects.Blob.from_string(content)
             node_path = node.name.encode(ENCODING)
             if dirnames:
                 # If there are trees which should be created we need to build
                 # them now (in reverse order)
                 reversed_dirnames = list(reversed(dirnames))
                 curtree = objects.Tree()
                 curtree[node_path] = node.mode, blob.id
                 new_trees.append(curtree)
                 for dirname in reversed_dirnames[:-1]:
                     newtree = objects.Tree()
                     #newtree.add(DIRMOD, dirname, curtree.id)
                     newtree[dirname] = DIRMOD, curtree.id
                     new_trees.append(newtree)
                     curtree = newtree
                 parent[reversed_dirnames[-1]] = DIRMOD, curtree.id
             else:
                 parent.add(name=node_path, mode=node.mode, hexsha=blob.id)
             new_trees.append(parent)
             # Update ancestors
             for parent, tree, path in reversed([(a[1], b[1], b[0]) for a, b in
                 zip(ancestors, ancestors[1:])]):
                 parent[path] = DIRMOD, tree.id
                 object_store.add_object(tree)
             object_store.add_object(blob)
             for tree in new_trees:
                 object_store.add_object(tree)
         for node in self.removed:
             paths = node.path.split('/')
             tree = commit_tree
             trees = [tree]
             # Traverse deep into the forest...
             for path in paths:
                 try:
                     obj = self.repository._repo[tree[path][1]]
                     if isinstance(obj, objects.Tree):
                         trees.append(obj)
                         tree = obj
                 except KeyError:
                     break
             # Cut down the blob and all rotten trees on the way back...
             for path, tree in reversed(zip(paths, trees)):
                 del tree[path]
                 if tree:
                     # This tree still has elements - don't remove it or any
                     # of it's parents
                     break
         object_store.add_object(commit_tree)
         # Create commit
         commit = objects.Commit()
         commit.tree = commit_tree.id
         commit.parents = [p._commit.id for p in self.parents if p]
         commit.author = commit.committer = safe_str(author)
         commit.encoding = ENCODING
         commit.message = safe_str(message)
         # Compute date
         if date is None:
             date = time.time()
         elif isinstance(date, datetime.datetime):
             date = time.mktime(date.timetuple())
         author_time = kwargs.pop('author_time', date)
         commit.commit_time = int(date)
         commit.author_time = int(author_time)
         tz = time.timezone
         author_tz = kwargs.pop('author_timezone', tz)
         commit.commit_timezone = tz
         commit.author_timezone = author_tz
         object_store.add_object(commit)
         ref = 'refs/heads/%s' % branch
         repo.refs[ref] = commit.id
         repo.refs.set_symbolic_ref('HEAD', ref)
         # Update vcs repository object & recreate dulwich repo
         self.repository.revisions.append(commit.id)
         # invalidate parsed refs after commit
         self.repository._parsed_refs = self.repository._get_parsed_refs()
         tip = self.repository.get_changeset()
         self.reset()
         return tip
     def _get_missing_trees(self, path, root_tree):
         """
         Creates missing ``Tree`` objects for the given path.
         :param path: path given as a string. It may be a path to a file node
           (i.e. ``foo/bar/baz.txt``) or directory path - in that case it must
           end with slash (i.e. ``foo/bar/``).
         :param root_tree: ``dulwich.objects.Tree`` object from which we start
           traversing (should be commit's root tree)
         """
         dirpath = posixpath.split(path)[0]
         dirs = dirpath.split('/')
         if not dirs or dirs == ['']:
             return []
         def get_tree_for_dir(tree, dirname):
             for name, mode, id in tree.iteritems():
                 if name == dirname:
                     obj = self.repository._repo[id]
                     if isinstance(obj, objects.Tree):
                         return obj
                     else:
                         raise RepositoryError("Cannot create directory %s "
                         "at tree %s as path is occupied and is not a "
                         "Tree" % (dirname, tree))
             return None
         trees = []
         parent = root_tree
         for dirname in dirs:
             tree = get_tree_for_dir(parent, dirname)
             if tree is None:
                 tree = objects.Tree()
                 dirmode = 040000
                 parent.add(dirmode, dirname, tree.id)
                 parent = tree
             # Always append tree
             trees.append(tree)
         return trees

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

➞

Show inline comments

 # -*- coding: utf-8 -*-
 """
     vcs.backends.git
     ~~~~~~~~~~~~~~~~
     vcs.backends.git.repository
     ~~~~~~~~~~~~~~~~~~~~~~~~~~~
-    Git backend implementation.
+    Git repository implementation.
     :created_on: Apr 8, 2010
     :copyright: (c) 2010-2011 by Marcin Kuzminski, Lukasz Balcerzak.
 """
 import os
 import re
 import time
 import posixpath
 import logging
 import traceback
 import urllib
 import urllib2
 from dulwich.repo import Repo, NotGitRepository
 import logging
 import posixpath
 import string
 from dulwich.objects import Tag
-from string import Template
+from dulwich.repo import Repo, NotGitRepository
-import rhodecode
+from rhodecode.lib.vcs import subprocessio
 from rhodecode.lib.vcs.backends.base import BaseRepository, CollectionGenerator
 from rhodecode.lib.vcs.exceptions import BranchDoesNotExistError
 from rhodecode.lib.vcs.exceptions import ChangesetDoesNotExistError
 from rhodecode.lib.vcs.exceptions import EmptyRepositoryError
 from rhodecode.lib.vcs.exceptions import RepositoryError
 from rhodecode.lib.vcs.exceptions import TagAlreadyExistError
 from rhodecode.lib.vcs.exceptions import TagDoesNotExistError
 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, ThreadLocalLazyProperty
 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.paths import get_user_home
 from .workdir import GitWorkdir
 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 .config import ConfigFile
 from rhodecode.lib import subprocessio
 from .workdir import GitWorkdir
 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 _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 = rhodecode.CONFIG.get('git_path', 'git')
+        _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
         """
         from mercurial.util import url as Url
         # those authnadlers are patched for python 2.6.5 bug an
         # infinit looping when given invalid resources
         from mercurial.url import httpbasicauthhandler, httpdigestauthhandler
         # 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 = Url(url).authinfo()
+        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 = rhodecode.CONFIG.get('git_rev_filter',
                                                       '--all').strip()
         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:
             raise EmptyRepositoryError("There are no changesets yet")
         if revision in (None, '', 'tip', 'HEAD', 'head', -1):
             revision = self.revisions[-1]
         if ((is_bstr(revision) 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):
             # 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]
             # maybe it's a tag ? we don't have them in self.revisions
             elif revision in _tags_shas:
                 return _tags_shas[_tags_shas.index(revision)]
             elif not 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)):
             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 <joe.doe@example.com>"
         :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 <joe.doe@example.com>"
         :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 = rhodecode.CONFIG.get('git_rev_filter',
                                                           '--all').strip()
             rev_filter = _git_path = settings.GIT_REV_FILTER
             cmd_template += ' %s' % (rev_filter)
         cmd = Template(cmd_template).safe_substitute(**cmd_params)
+        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/changeset.py

➞

Show inline comments

 import os
 import posixpath
 from rhodecode.lib.vcs.conf import settings
 from rhodecode.lib.vcs.backends.base import BaseChangeset
 from rhodecode.lib.vcs.conf import settings
 from rhodecode.lib.vcs.exceptions import  ChangesetDoesNotExistError, \
     ChangesetError, ImproperArchiveTypeError, NodeDoesNotExistError, VCSError
 from rhodecode.lib.vcs.nodes import AddedFileNodesGenerator, \
     ChangedFileNodesGenerator, DirNode, FileNode, NodeKind, \
     RemovedFileNodesGenerator, RootNode, SubModuleNode
 from rhodecode.lib.vcs.exceptions import (
     ChangesetDoesNotExistError, ChangesetError, ImproperArchiveTypeError,
     NodeDoesNotExistError, VCSError
+)
 from rhodecode.lib.vcs.nodes import (
     AddedFileNodesGenerator, ChangedFileNodesGenerator, DirNode, FileNode,
     NodeKind, RemovedFileNodesGenerator, RootNode, SubModuleNode
+)
 from rhodecode.lib.vcs.utils import safe_str, safe_unicode, date_fromtimestamp
 from rhodecode.lib.vcs.utils.lazy import LazyProperty
 from rhodecode.lib.vcs.utils.paths import get_dirs_for_path
 from rhodecode.lib.vcs.utils.hgcompat import archival, hex
 class MercurialChangeset(BaseChangeset):
     """
     Represents state of the repository at the single revision.
     """
     def __init__(self, repository, revision):
         self.repository = repository
         self.raw_id = revision
         self._ctx = repository._repo[revision]
         self.revision = self._ctx._rev
         self.nodes = {}
     @LazyProperty
     def tags(self):
         return map(safe_unicode, self._ctx.tags())
     @LazyProperty
     def branch(self):
         return  safe_unicode(self._ctx.branch())
     @LazyProperty
     def bookmarks(self):
         return map(safe_unicode, self._ctx.bookmarks())
     @LazyProperty
     def message(self):
         return safe_unicode(self._ctx.description())
     @LazyProperty
     def committer(self):
         return safe_unicode(self.author)
     @LazyProperty
     def author(self):
         return safe_unicode(self._ctx.user())
     @LazyProperty
     def date(self):
         return date_fromtimestamp(*self._ctx.date())
     @LazyProperty
     def _timestamp(self):
         return self._ctx.date()[0]
     @LazyProperty
     def status(self):
         """
         Returns modified, added, removed, deleted files for current changeset
         """
         return self.repository._repo.status(self._ctx.p1().node(),
                                             self._ctx.node())
     @LazyProperty
     def _file_paths(self):
         return list(self._ctx)
     @LazyProperty
     def _dir_paths(self):
         p = list(set(get_dirs_for_path(*self._file_paths)))
         p.insert(0, '')
         return p
     @LazyProperty
     def _paths(self):
         return self._dir_paths + self._file_paths
     @LazyProperty
     def id(self):
         if self.last:
             return u'tip'
         return self.short_id
     @LazyProperty
     def short_id(self):
         return self.raw_id[:12]
     @LazyProperty
     def parents(self):
         """
         Returns list of parents changesets.
         """
         return [self.repository.get_changeset(parent.rev())
                 for parent in self._ctx.parents() if parent.rev() >= 0]
     @LazyProperty
     def children(self):
         """
         Returns list of children changesets.
         """
         return [self.repository.get_changeset(child.rev())
                 for child in self._ctx.children() if child.rev() >= 0]
     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):
         return ''.join(self._ctx.diff(git=True,
                                       ignore_whitespace=ignore_whitespace,
                                       context=context))
     def _fix_path(self, path):
         """
         Paths are stored without trailing slash so we need to get rid off it if
         needed. Also mercurial keeps filenodes as str so we need to decode
         from unicode to str
         """
         if path.endswith('/'):
             path = path.rstrip('/')
         return safe_str(path)
     def _get_kind(self, path):
         path = self._fix_path(path)
         if path in self._file_paths:
             return NodeKind.FILE
         elif path in self._dir_paths:
             return NodeKind.DIR
         else:
             raise ChangesetError("Node does not exist at the given path '%s'"
                 % (path))
     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 self._ctx.filectx(path)
     def _extract_submodules(self):
         """
         returns a dictionary with submodule information from substate file
         of hg repository
         """
         return self._ctx.substate
     def get_file_mode(self, path):
         """
         Returns stat mode of the file at the given ``path``.
         """
         fctx = self._get_filectx(path)
         if 'x' in fctx.flags():
             return 0100755
         else:
             return 0100644
     def get_file_content(self, path):
         """
         Returns content of the file at given ``path``.
         """
         fctx = self._get_filectx(path)
         return fctx.data()
     def get_file_size(self, path):
         """
         Returns size of the file at given ``path``.
         """
         fctx = self._get_filectx(path)
         return fctx.size()
     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.
         """
         fctx = self._get_filectx(path)
         hist = []
         cnt = 0
         for cs in reversed([x for x in fctx.filelog()]):
             cnt += 1
             hist.append(hex(fctx.filectx(cs).node()))
             if limit and cnt == limit:
                 break
         return [self.repository.get_changeset(node) for node in hist]
     def get_file_annotate(self, path):
         """
         Returns a generator of four element tuples with
             lineno, sha, changeset lazy loader and line
         """
         fctx = self._get_filectx(path)
         for i, annotate_data in enumerate(fctx.annotate()):
             ln_no = i + 1
             sha = hex(annotate_data[0].node())
             yield (ln_no, sha, lambda: self.repository.get_changeset(sha), annotate_data[1],)
     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 stream is None:
             raise VCSError('You need to pass in a valid stream for filling'
                            ' with archival data')
         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")
         archival.archive(self.repository._repo, stream, self.raw_id,
                          kind, prefix=prefix, subrepos=subrepos)
         if stream.closed and hasattr(stream, 'name'):
             stream = open(stream.name, 'rb')
         elif hasattr(stream, 'mode') and 'r' not in stream.mode:
             stream = open(stream.name, 'rb')
         else:
             stream.seek(0)
     def get_nodes(self, path):
         """
         Returns combined ``DirNode`` and ``FileNode`` objects list representing
         state of changeset at the given ``path``. If node at the given ``path``
         is not instance of ``DirNode``, ChangesetError would be raised.
         """
         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)
         filenodes = [FileNode(f, changeset=self) for f in self._file_paths
             if os.path.dirname(f) == path]
         dirs = path == '' and '' or [d for d in self._dir_paths
             if d and posixpath.dirname(d) == path]
         dirnodes = [DirNode(d, changeset=self) for d in dirs
             if os.path.dirname(d) == path]
         als = self.repository.alias
         for k, vals in self._extract_submodules().iteritems():
             #vals = url,rev,type
             loc = vals[0]
             cs = vals[1]
             dirnodes.append(SubModuleNode(k, url=loc, changeset=cs,
                                           alias=als))
         nodes = dirnodes + filenodes
         # cache nodes
         for node in nodes:
             self.nodes[node.path] = node
         nodes.sort()
         return nodes
     def get_node(self, path):
         """
         Returns ``Node`` object from the given ``path``. If there is no node at
         the given ``path``, ``ChangesetError`` would be raised.
         """
         path = self._fix_path(path)
         if not path in self.nodes:
             if path in self._file_paths:
                 node = FileNode(path, changeset=self)
             elif path in self._dir_paths or path in self._dir_paths:
                 if path == '':
                     node = RootNode(changeset=self)
                 else:
                     node = DirNode(path, changeset=self)
             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
         """
         return self._ctx.files()
     @property
     def added(self):
         """
         Returns list of added ``FileNode`` objects.
         """
         return AddedFileNodesGenerator([n for n in self.status[1]], self)
     @property
     def changed(self):
         """
         Returns list of modified ``FileNode`` objects.
         """
         return ChangedFileNodesGenerator([n for n in  self.status[0]], self)
     @property
     def removed(self):
         """
         Returns list of removed ``FileNode`` objects.
         """
         return RemovedFileNodesGenerator([n for n in self.status[2]], self)

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

➞

Show inline comments

 # -*- 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 datetime
 import urllib
 import urllib2
 import logging
 import datetime
 from rhodecode.lib.vcs.backends.base import BaseRepository, CollectionGenerator
 from .workdir import MercurialWorkdir
 from .changeset import MercurialChangeset
 from .inmemory import MercurialInMemoryChangeset
 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.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 rhodecode.lib.vcs.utils.hgcompat import ui, nullid, match, patch, \
     diffopts, clone, get_contact, pull, localrepository, RepoLookupError, \
     Abort, RepoError, hex, scmutil
 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 <joe.doe@example.com>"
         :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 <joe.doe@example.com>"
         :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
         """
         from mercurial.util import url as Url
         # those authnadlers are patched for python 2.6.5 bug an
         # infinit looping when given invalid resources
         from mercurial.url import httpbasicauthhandler, httpdigestauthhandler
         # 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 = Url(url).authinfo()
+        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
 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:
             filter_.append('date(">%s")' % start_date)
         if end_date:
             filter_.append('date("<%s")' % end_date)
         if filter_:
             revisions = scmutil.revrange(self._repo, filter_)
         else:
             revisions = self.revisions
         revs = reversed(revisions[start_pos:end_pos]) if reverse else \
                 revisions[start_pos:end_pos]
         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/lib/vcs/conf/settings.py

➞

Show inline comments

 import os
 import tempfile
 from rhodecode.lib.vcs.utils import aslist
 from rhodecode.lib.vcs.utils.paths import get_user_home
 abspath = lambda * p: os.path.abspath(os.path.join(*p))
 VCSRC_PATH = os.environ.get('VCSRC_PATH')
 if not VCSRC_PATH:
     HOME_ = get_user_home()
     if not HOME_:
         HOME_ = tempfile.gettempdir()
 VCSRC_PATH = VCSRC_PATH or abspath(HOME_, '.vcsrc')
 if os.path.isdir(VCSRC_PATH):
     VCSRC_PATH = os.path.join(VCSRC_PATH, '__init__.py')
 # list of default encoding used in safe_unicode/safe_str methods
 DEFAULT_ENCODINGS = aslist('utf8')
 # path to git executable runned by run_git_command function
 GIT_EXECUTABLE_PATH = 'git'
 # can be also --branches --tags
 GIT_REV_FILTER = '--all'
 BACKENDS = {
     'hg': 'vcs.backends.hg.MercurialRepository',
     'git': 'vcs.backends.git.GitRepository',
     'hg': 'rhodecode.lib.vcs.backends.hg.MercurialRepository',
     'git': 'rhodecode.lib.vcs.backends.git.GitRepository',
+}
 ARCHIVE_SPECS = {
     'tar': ('application/x-tar', '.tar'),
     'tbz2': ('application/x-bzip2', '.tar.bz2'),
     'tgz': ('application/x-gzip', '.tar.gz'),
     'zip': ('application/zip', '.zip'),
+}
 BACKENDS = {
     'hg': 'rhodecode.lib.vcs.backends.hg.MercurialRepository',
     'git': 'rhodecode.lib.vcs.backends.git.GitRepository',
+}

rhodecode/lib/vcs/nodes.py

➞

Show inline comments

 # -*- coding: utf-8 -*-
 """
     vcs.nodes
     ~~~~~~~~~
     Module holding everything related to vcs nodes.
     :created_on: Apr 8, 2010
     :copyright: (c) 2010-2011 by Marcin Kuzminski, Lukasz Balcerzak.
 """
 import os
 import stat
 import posixpath
 import mimetypes
 from pygments import lexers
 from rhodecode.lib.vcs.backends.base import EmptyChangeset
 from rhodecode.lib.vcs.exceptions import NodeError, RemovedFileNodeError
 from rhodecode.lib.vcs.utils.lazy import LazyProperty
 from rhodecode.lib.vcs.utils import safe_unicode
 from rhodecode.lib.vcs.exceptions import NodeError
 from rhodecode.lib.vcs.exceptions import RemovedFileNodeError
 from rhodecode.lib.vcs.backends.base import EmptyChangeset
 class NodeKind:
     SUBMODULE = -1
     DIR = 1
     FILE = 2
 class NodeState:
     ADDED = u'added'
     CHANGED = u'changed'
     NOT_CHANGED = u'not changed'
     REMOVED = u'removed'
 class NodeGeneratorBase(object):
     """
     Base class for removed added and changed filenodes, it's a lazy generator
     class that will create filenodes only on iteration or call
     The len method doesn't need to create filenodes at all
     """
     def __init__(self, current_paths, cs):
         self.cs = cs
         self.current_paths = current_paths
     def __call__(self):
         return [n for n in self]
     def __getslice__(self, i, j):
         for p in self.current_paths[i:j]:
             yield self.cs.get_node(p)
     def __len__(self):
         return len(self.current_paths)
     def __iter__(self):
         for p in self.current_paths:
             yield self.cs.get_node(p)
 class AddedFileNodesGenerator(NodeGeneratorBase):
     """
     Class holding Added files for current changeset
     """
     pass
 class ChangedFileNodesGenerator(NodeGeneratorBase):
     """
     Class holding Changed files for current changeset
     """
     pass
 class RemovedFileNodesGenerator(NodeGeneratorBase):
     """
     Class holding removed files for current changeset
     """
     def __iter__(self):
         for p in self.current_paths:
             yield RemovedFileNode(path=p)
     def __getslice__(self, i, j):
         for p in self.current_paths[i:j]:
             yield RemovedFileNode(path=p)
 class Node(object):
     """
     Simplest class representing file or directory on repository.  SCM backends
     should use ``FileNode`` and ``DirNode`` subclasses rather than ``Node``
     directly.
     Node's ``path`` cannot start with slash as we operate on *relative* paths
     only. Moreover, every single node is identified by the ``path`` attribute,
     so it cannot end with slash, too. Otherwise, path could lead to mistakes.
     """
     def __init__(self, path, kind):
         if path.startswith('/'):
             raise NodeError("Cannot initialize Node objects with slash at "
                 "the beginning as only relative paths are supported")
         self.path = path.rstrip('/')
         if path == '' and kind != NodeKind.DIR:
             raise NodeError("Only DirNode and its subclasses may be "
                             "initialized with empty path")
         self.kind = kind
         #self.dirs, self.files = [], []
         if self.is_root() and not self.is_dir():
             raise NodeError("Root node cannot be FILE kind")
     @LazyProperty
     def parent(self):
         parent_path = self.get_parent_path()
         if parent_path:
             if self.changeset:
                 return self.changeset.get_node(parent_path)
             return DirNode(parent_path)
         return None
     @LazyProperty
     def unicode_path(self):
         return safe_unicode(self.path)
     @LazyProperty
     def name(self):
         """
         Returns name of the node so if its path
         then only last part is returned.
         """
         return safe_unicode(self.path.rstrip('/').split('/')[-1])
     def _get_kind(self):
         return self._kind
     def _set_kind(self, kind):
         if hasattr(self, '_kind'):
             raise NodeError("Cannot change node's kind")
         else:
             self._kind = kind
             # Post setter check (path's trailing slash)
             if self.path.endswith('/'):
                 raise NodeError("Node's path cannot end with slash")
     kind = property(_get_kind, _set_kind)
     def __cmp__(self, other):
         """
         Comparator using name of the node, needed for quick list sorting.
         """
         kind_cmp = cmp(self.kind, other.kind)
         if kind_cmp:
             return kind_cmp
         return cmp(self.name, other.name)
     def __eq__(self, other):
         for attr in ['name', 'path', 'kind']:
             if getattr(self, attr) != getattr(other, attr):
                 return False
         if self.is_file():
             if self.content != other.content:
                 return False
         else:
             # For DirNode's check without entering each dir
             self_nodes_paths = list(sorted(n.path for n in self.nodes))
             other_nodes_paths = list(sorted(n.path for n in self.nodes))
             if self_nodes_paths != other_nodes_paths:
                 return False
         return True
     def __nq__(self, other):
         return not self.__eq__(other)
     def __repr__(self):
         return '<%s %r>' % (self.__class__.__name__, self.path)
     def __str__(self):
         return self.__repr__()
     def __unicode__(self):
         return self.name
     def get_parent_path(self):
         """
         Returns node's parent path or empty string if node is root.
         """
         if self.is_root():
             return ''
         return posixpath.dirname(self.path.rstrip('/')) + '/'
     def is_file(self):
         """
         Returns ``True`` if node's kind is ``NodeKind.FILE``, ``False``
         otherwise.
         """
         return self.kind == NodeKind.FILE
     def is_dir(self):
         """
         Returns ``True`` if node's kind is ``NodeKind.DIR``, ``False``
         otherwise.
         """
         return self.kind == NodeKind.DIR
     def is_root(self):
         """
         Returns ``True`` if node is a root node and ``False`` otherwise.
         """
         return self.kind == NodeKind.DIR and self.path == ''
     def is_submodule(self):
         """
         Returns ``True`` if node's kind is ``NodeKind.SUBMODULE``, ``False``
         otherwise.
         """
         return self.kind == NodeKind.SUBMODULE
     @LazyProperty
     def added(self):
         return self.state is NodeState.ADDED
     @LazyProperty
     def changed(self):
         return self.state is NodeState.CHANGED
     @LazyProperty
     def not_changed(self):
         return self.state is NodeState.NOT_CHANGED
     @LazyProperty
     def removed(self):
         return self.state is NodeState.REMOVED
 class FileNode(Node):
     """
     Class representing file nodes.
     :attribute: path: path to the node, relative to repostiory's root
     :attribute: content: if given arbitrary sets content of the file
     :attribute: changeset: if given, first time content is accessed, callback
     :attribute: mode: octal stat mode for a node. Default is 0100644.
     """
     def __init__(self, path, content=None, changeset=None, mode=None):
         """
         Only one of ``content`` and ``changeset`` may be given. Passing both
         would raise ``NodeError`` exception.
         :param path: relative path to the node
         :param content: content may be passed to constructor
         :param changeset: if given, will use it to lazily fetch content
         :param mode: octal representation of ST_MODE (i.e. 0100644)
         """
         if content and changeset:
             raise NodeError("Cannot use both content and changeset")
         super(FileNode, self).__init__(path, kind=NodeKind.FILE)
         self.changeset = changeset
         self._content = content
         self._mode = mode or 0100644
     @LazyProperty
     def mode(self):
         """
         Returns lazily mode of the FileNode. If ``changeset`` is not set, would
         use value given at initialization or 0100644 (default).
         """
         if self.changeset:
             mode = self.changeset.get_file_mode(self.path)
         else:
             mode = self._mode
         return mode
     def _get_content(self):
         if self.changeset:
             content = self.changeset.get_file_content(self.path)
         else:
             content = self._content
         return content
     @property
     def content(self):
         """
         Returns lazily content of the FileNode. If possible, would try to
         decode content from UTF-8.
         """
         content = self._get_content()
         if bool(content and '\0' in content):
             return content
         return safe_unicode(content)
     @LazyProperty
     def size(self):
         if self.changeset:
             return self.changeset.get_file_size(self.path)
         raise NodeError("Cannot retrieve size of the file without related "
             "changeset attribute")
     @LazyProperty
     def message(self):
         if self.changeset:
             return self.last_changeset.message
         raise NodeError("Cannot retrieve message of the file without related "
             "changeset attribute")
     @LazyProperty
     def last_changeset(self):
         if self.changeset:
             return self.changeset.get_file_changeset(self.path)
         raise NodeError("Cannot retrieve last changeset of the file without "
             "related changeset attribute")
     def get_mimetype(self):
         """
         Mimetype is calculated based on the file's content. If ``_mimetype``
         attribute is available, it will be returned (backends which store
         mimetypes or can easily recognize them, should set this private
         attribute to indicate that type should *NOT* be calculated).
         """
         if hasattr(self, '_mimetype'):
             if (isinstance(self._mimetype, (tuple, list,)) and
                 len(self._mimetype) == 2):
                 return self._mimetype
             else:
                 raise NodeError('given _mimetype attribute must be an 2 '
                                'element list or tuple')
         mtype, encoding = mimetypes.guess_type(self.name)
         if mtype is None:
             if self.is_binary:
                 mtype = 'application/octet-stream'
                 encoding = None
             else:
                 mtype = 'text/plain'
                 encoding = None
         return mtype, encoding
     @LazyProperty
     def mimetype(self):
         """
         Wrapper around full mimetype info. It returns only type of fetched
         mimetype without the encoding part. use get_mimetype function to fetch
         full set of (type,encoding)
         """
         return self.get_mimetype()[0]
     @LazyProperty
     def mimetype_main(self):
         return ['', '']
         return self.mimetype.split('/')[0]
     @LazyProperty
     def lexer(self):
         """
         Returns pygment's lexer class. Would try to guess lexer taking file's
         content, name and mimetype.
         """
         try:
             lexer = lexers.guess_lexer_for_filename(self.name, self.content, stripnl=False)
         except lexers.ClassNotFound:
             lexer = lexers.TextLexer(stripnl=False)
         # returns first alias
         return lexer
     @LazyProperty
     def lexer_alias(self):
         """
         Returns first alias of the lexer guessed for this file.
         """
         return self.lexer.aliases[0]
     @LazyProperty
     def history(self):
         """
         Returns a list of changeset for this file in which the file was changed
         """
         if self.changeset is None:
             raise NodeError('Unable to get changeset for this FileNode')
         return self.changeset.get_file_history(self.path)
     @LazyProperty
     def annotate(self):
         """
         Returns a list of three element tuples with lineno,changeset and line
         """
         if self.changeset is None:
             raise NodeError('Unable to get changeset for this FileNode')
         return self.changeset.get_file_annotate(self.path)
     @LazyProperty
     def state(self):
         if not self.changeset:
             raise NodeError("Cannot check state of the node if it's not "
                 "linked with changeset")
         elif self.path in (node.path for node in self.changeset.added):
             return NodeState.ADDED
         elif self.path in (node.path for node in self.changeset.changed):
             return NodeState.CHANGED
         else:
             return NodeState.NOT_CHANGED
     @property
     def is_binary(self):
         """
         Returns True if file has binary content.
         """
         _bin = '\0' in self._get_content()
         return _bin
     @LazyProperty
     def extension(self):
         """Returns filenode extension"""
         return self.name.split('.')[-1]
     def is_executable(self):
         """
         Returns ``True`` if file has executable flag turned on.
         """
         return bool(self.mode & stat.S_IXUSR)
     def __repr__(self):
         return '<%s %r @ %s>' % (self.__class__.__name__, self.path,
                                  getattr(self.changeset, 'short_id', ''))
 class RemovedFileNode(FileNode):
     """
     Dummy FileNode class - trying to access any public attribute except path,
     name, kind or state (or methods/attributes checking those two) would raise
     RemovedFileNodeError.
     """
     ALLOWED_ATTRIBUTES = [
         'name', 'path', 'state', 'is_root', 'is_file', 'is_dir', 'kind',
         'added', 'changed', 'not_changed', 'removed'
+    ]
     def __init__(self, path):
         """
         :param path: relative path to the node
         """
         super(RemovedFileNode, self).__init__(path=path)
     def __getattribute__(self, attr):
         if attr.startswith('_') or attr in RemovedFileNode.ALLOWED_ATTRIBUTES:
             return super(RemovedFileNode, self).__getattribute__(attr)
         raise RemovedFileNodeError("Cannot access attribute %s on "
             "RemovedFileNode" % attr)
     @LazyProperty
     def state(self):
         return NodeState.REMOVED
 class DirNode(Node):
     """
     DirNode stores list of files and directories within this node.
     Nodes may be used standalone but within repository context they
     lazily fetch data within same repositorty's changeset.
     """
     def __init__(self, path, nodes=(), changeset=None):
         """
         Only one of ``nodes`` and ``changeset`` may be given. Passing both
         would raise ``NodeError`` exception.
         :param path: relative path to the node
         :param nodes: content may be passed to constructor
         :param changeset: if given, will use it to lazily fetch content
         :param size: always 0 for ``DirNode``
         """
         if nodes and changeset:
             raise NodeError("Cannot use both nodes and changeset")
         super(DirNode, self).__init__(path, NodeKind.DIR)
         self.changeset = changeset
         self._nodes = nodes
     @LazyProperty
     def content(self):
         raise NodeError("%s represents a dir and has no ``content`` attribute"
             % self)
     @LazyProperty
     def nodes(self):
         if self.changeset:
             nodes = self.changeset.get_nodes(self.path)
         else:
             nodes = self._nodes
         self._nodes_dict = dict((node.path, node) for node in nodes)
         return sorted(nodes)
     @LazyProperty
     def files(self):
         return sorted((node for node in self.nodes if node.is_file()))
     @LazyProperty
     def dirs(self):
         return sorted((node for node in self.nodes if node.is_dir()))
     def __iter__(self):
         for node in self.nodes:
             yield node
     def get_node(self, path):
         """
         Returns node from within this particular ``DirNode``, so it is now
         allowed to fetch, i.e. node located at 'docs/api/index.rst' from node
         'docs'. In order to access deeper nodes one must fetch nodes between
         them first - this would work::
            docs = root.get_node('docs')
            docs.get_node('api').get_node('index.rst')
         :param: path - relative to the current node
         .. note::
            To access lazily (as in example above) node have to be initialized
            with related changeset object - without it node is out of
            context and may know nothing about anything else than nearest
            (located at same level) nodes.
         """
         try:
             path = path.rstrip('/')
             if path == '':
                 raise NodeError("Cannot retrieve node without path")
             self.nodes  # access nodes first in order to set _nodes_dict
             paths = path.split('/')
             if len(paths) == 1:
                 if not self.is_root():
                     path = '/'.join((self.path, paths[0]))
                 else:
                     path = paths[0]
                 return self._nodes_dict[path]
             elif len(paths) > 1:
                 if self.changeset is None:
                     raise NodeError("Cannot access deeper "
                                     "nodes without changeset")
                 else:
                     path1, path2 = paths[0], '/'.join(paths[1:])
                     return self.get_node(path1).get_node(path2)
             else:
                 raise KeyError
         except KeyError:
             raise NodeError("Node does not exist at %s" % path)
     @LazyProperty
     def state(self):
         raise NodeError("Cannot access state of DirNode")
     @LazyProperty
     def size(self):
         size = 0
         for root, dirs, files in self.changeset.walk(self.path):
             for f in files:
                 size += f.size
         return size
     def __repr__(self):
         return '<%s %r @ %s>' % (self.__class__.__name__, self.path,
                                  getattr(self.changeset, 'short_id', ''))
 class RootNode(DirNode):
     """
     DirNode being the root node of the repository.
     """
     def __init__(self, nodes=(), changeset=None):
         super(RootNode, self).__init__(path='', nodes=nodes,
             changeset=changeset)
     def __repr__(self):
         return '<%s>' % self.__class__.__name__
 class SubModuleNode(Node):
     """
     represents a SubModule of Git or SubRepo of Mercurial
     """
     is_binary = False
     size = 0
     def __init__(self, name, url=None, changeset=None, alias=None):
         self.path = name
         self.kind = NodeKind.SUBMODULE
         self.alias = alias
         # we have to use emptyChangeset here since this can point to svn/git/hg
         # submodules we cannot get from repository
         self.changeset = EmptyChangeset(str(changeset), alias=alias)
         self.url = url or self._extract_submodule_url()
     def __repr__(self):
         return '<%s %r @ %s>' % (self.__class__.__name__, self.path,
                                  getattr(self.changeset, 'short_id', ''))
     def _extract_submodule_url(self):
         if self.alias == 'git':
             #TODO: find a way to parse gits submodule file and extract the
             # linking URL
             return self.path
         if self.alias == 'hg':
             return self.path
     @LazyProperty
     def name(self):
         """
         Returns name of the node so if its path
         then only last part is returned.
         """
         org = safe_unicode(self.path.rstrip('/').split('/')[-1])
         return u'%s @ %s' % (org, self.changeset.short_id)

rhodecode/lib/vcs/subprocessio.py

➞

Show inline comments

@@ new file 100644 @@
 '''
 Module provides a class allowing to wrap communication over subprocess.Popen
 input, output, error streams into a meaningfull, non-blocking, concurrent
 stream processor exposing the output data as an iterator fitting to be a
 return value passed by a WSGI applicaiton to a WSGI server per PEP 3333.
 Copyright (c) 2011  Daniel Dotsenko <dotsa@hotmail.com>
 This file is part of git_http_backend.py Project.
 git_http_backend.py Project is free software: you can redistribute it and/or
 modify it under the terms of the GNU Lesser General Public License as
 published by the Free Software Foundation, either version 2.1 of the License,
 or (at your option) any later version.
 git_http_backend.py Project is distributed in the hope that it will be useful,
 but WITHOUT ANY WARRANTY; without even the implied warranty of
 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
 GNU Lesser General Public License for more details.
 You should have received a copy of the GNU Lesser General Public License
 along with git_http_backend.py Project.
 If not, see <http://www.gnu.org/licenses/>.
 '''
 import os
 import subprocess
 from rhodecode.lib.vcs.utils.compat import deque, Event, Thread, _bytes, _bytearray
 class StreamFeeder(Thread):
     """
     Normal writing into pipe-like is blocking once the buffer is filled.
     This thread allows a thread to seep data from a file-like into a pipe
     without blocking the main thread.
     We close inpipe once the end of the source stream is reached.
     """
     def __init__(self, source):
         super(StreamFeeder, self).__init__()
         self.daemon = True
         filelike = False
         self.bytes = _bytes()
         if type(source) in (type(''), _bytes, _bytearray):  # string-like
             self.bytes = _bytes(source)
         else:  # can be either file pointer or file-like
             if type(source) in (int, long):  # file pointer it is
                 ## converting file descriptor (int) stdin into file-like
                 try:
                     source = os.fdopen(source, 'rb', 16384)
                 except Exception:
                     pass
             # let's see if source is file-like by now
             try:
                 filelike = source.read
             except Exception:
                 pass
         if not filelike and not self.bytes:
             raise TypeError("StreamFeeder's source object must be a readable "
                             "file-like, a file descriptor, or a string-like.")
         self.source = source
         self.readiface, self.writeiface = os.pipe()
     def run(self):
         t = self.writeiface
         if self.bytes:
             os.write(t, self.bytes)
         else:
             s = self.source
             b = s.read(4096)
             while b:
                 os.write(t, b)
                 b = s.read(4096)
         os.close(t)
     @property
     def output(self):
         return self.readiface
 class InputStreamChunker(Thread):
     def __init__(self, source, target, buffer_size, chunk_size):
         super(InputStreamChunker, self).__init__()
         self.daemon = True  # die die die.
         self.source = source
         self.target = target
         self.chunk_count_max = int(buffer_size / chunk_size) + 1
         self.chunk_size = chunk_size
         self.data_added = Event()
         self.data_added.clear()
         self.keep_reading = Event()
         self.keep_reading.set()
         self.EOF = Event()
         self.EOF.clear()
         self.go = Event()
         self.go.set()
     def stop(self):
         self.go.clear()
         self.EOF.set()
         try:
             # this is not proper, but is done to force the reader thread let
             # go of the input because, if successful, .close() will send EOF
             # down the pipe.
             self.source.close()
         except:
             pass
     def run(self):
         s = self.source
         t = self.target
         cs = self.chunk_size
         ccm = self.chunk_count_max
         kr = self.keep_reading
         da = self.data_added
         go = self.go
         try:
             b = s.read(cs)
         except ValueError:
             b = ''
         while b and go.is_set():
             if len(t) > ccm:
                 kr.clear()
                 kr.wait(2)
 #                # this only works on 2.7.x and up
 #                if not kr.wait(10):
 #                    raise Exception("Timed out while waiting for input to be read.")
                 # instead we'll use this
                 if len(t) > ccm + 3:
                     raise IOError("Timed out while waiting for input from subprocess.")
             t.append(b)
             da.set()
             b = s.read(cs)
         self.EOF.set()
         da.set()  # for cases when done but there was no input.
 class BufferedGenerator():
     '''
     Class behaves as a non-blocking, buffered pipe reader.
     Reads chunks of data (through a thread)
     from a blocking pipe, and attaches these to an array (Deque) of chunks.
     Reading is halted in the thread when max chunks is internally buffered.
     The .next() may operate in blocking or non-blocking fashion by yielding
     '' if no data is ready
     to be sent or by not returning until there is some data to send
     When we get EOF from underlying source pipe we raise the marker to raise
     StopIteration after the last chunk of data is yielded.
     '''
     def __init__(self, source, buffer_size=65536, chunk_size=4096,
                  starting_values=[], bottomless=False):
         if bottomless:
             maxlen = int(buffer_size / chunk_size)
         else:
             maxlen = None
         self.data = deque(starting_values, maxlen)
         self.worker = InputStreamChunker(source, self.data, buffer_size,
                                          chunk_size)
         if starting_values:
             self.worker.data_added.set()
         self.worker.start()
     ####################
     # Generator's methods
     ####################
     def __iter__(self):
         return self
     def next(self):
         while not len(self.data) and not self.worker.EOF.is_set():
             self.worker.data_added.clear()
             self.worker.data_added.wait(0.2)
         if len(self.data):
             self.worker.keep_reading.set()
             return _bytes(self.data.popleft())
         elif self.worker.EOF.is_set():
             raise StopIteration
     def throw(self, type, value=None, traceback=None):
         if not self.worker.EOF.is_set():
             raise type(value)
     def start(self):
         self.worker.start()
     def stop(self):
         self.worker.stop()
     def close(self):
         try:
             self.worker.stop()
             self.throw(GeneratorExit)
         except (GeneratorExit, StopIteration):
             pass
     def __del__(self):
         self.close()
     ####################
     # Threaded reader's infrastructure.
     ####################
     @property
     def input(self):
         return self.worker.w
     @property
     def data_added_event(self):
         return self.worker.data_added
     @property
     def data_added(self):
         return self.worker.data_added.is_set()
     @property
     def reading_paused(self):
         return not self.worker.keep_reading.is_set()
     @property
     def done_reading_event(self):
         '''
         Done_reding does not mean that the iterator's buffer is empty.
         Iterator might have done reading from underlying source, but the read
         chunks might still be available for serving through .next() method.
         @return An Event class instance.
         '''
         return self.worker.EOF
     @property
     def done_reading(self):
         '''
         Done_reding does not mean that the iterator's buffer is empty.
         Iterator might have done reading from underlying source, but the read
         chunks might still be available for serving through .next() method.
         @return An Bool value.
         '''
         return self.worker.EOF.is_set()
     @property
     def length(self):
         '''
         returns int.
         This is the lenght of the que of chunks, not the length of
         the combined contents in those chunks.
         __len__() cannot be meaningfully implemented because this
         reader is just flying throuh a bottomless pit content and
         can only know the lenght of what it already saw.
         If __len__() on WSGI server per PEP 3333 returns a value,
         the responce's length will be set to that. In order not to
         confuse WSGI PEP3333 servers, we will not implement __len__
         at all.
         '''
         return len(self.data)
     def prepend(self, x):
         self.data.appendleft(x)
     def append(self, x):
         self.data.append(x)
     def extend(self, o):
         self.data.extend(o)
     def __getitem__(self, i):
         return self.data[i]
 class SubprocessIOChunker(object):
     '''
     Processor class wrapping handling of subprocess IO.
     In a way, this is a "communicate()" replacement with a twist.
     - We are multithreaded. Writing in and reading out, err are all sep threads.
     - We support concurrent (in and out) stream processing.
     - The output is not a stream. It's a queue of read string (bytes, not unicode)
       chunks. The object behaves as an iterable. You can "for chunk in obj:" us.
     - We are non-blocking in more respects than communicate()
       (reading from subprocess out pauses when internal buffer is full, but
        does not block the parent calling code. On the flip side, reading from
        slow-yielding subprocess may block the iteration until data shows up. This
        does not block the parallel inpipe reading occurring parallel thread.)
     The purpose of the object is to allow us to wrap subprocess interactions into
     and interable that can be passed to a WSGI server as the application's return
     value. Because of stream-processing-ability, WSGI does not have to read ALL
     of the subprocess's output and buffer it, before handing it to WSGI server for
     HTTP response. Instead, the class initializer reads just a bit of the stream
     to figure out if error ocurred or likely to occur and if not, just hands the
     further iteration over subprocess output to the server for completion of HTTP
     response.
     The real or perceived subprocess error is trapped and raised as one of
     EnvironmentError family of exceptions
     Example usage:
     #    try:
     #        answer = SubprocessIOChunker(
     #            cmd,
     #            input,
     #            buffer_size = 65536,
     #            chunk_size = 4096
     #            )
     #    except (EnvironmentError) as e:
     #        print str(e)
     #        raise e
+    #
     #    return answer
     '''
     def __init__(self, cmd, inputstream=None, buffer_size=65536,
                  chunk_size=4096, starting_values=[], **kwargs):
         '''
         Initializes SubprocessIOChunker
         :param cmd: A Subprocess.Popen style "cmd". Can be string or array of strings
         :param inputstream: (Default: None) A file-like, string, or file pointer.
         :param buffer_size: (Default: 65536) A size of total buffer per stream in bytes.
         :param chunk_size: (Default: 4096) A max size of a chunk. Actual chunk may be smaller.
         :param starting_values: (Default: []) An array of strings to put in front of output que.
         '''
         if inputstream:
             input_streamer = StreamFeeder(inputstream)
             input_streamer.start()
             inputstream = input_streamer.output
         if isinstance(cmd, (list, tuple)):
             cmd = ' '.join(cmd)
         _shell = kwargs.get('shell') or True
         kwargs['shell'] = _shell
         _p = subprocess.Popen(cmd,
             bufsize=-1,
             stdin=inputstream,
             stdout=subprocess.PIPE,
             stderr=subprocess.PIPE,
             **kwargs
+            )
         bg_out = BufferedGenerator(_p.stdout, buffer_size, chunk_size, starting_values)
         bg_err = BufferedGenerator(_p.stderr, 16000, 1, bottomless=True)
         while not bg_out.done_reading and not bg_out.reading_paused and not bg_err.length:
             # doing this until we reach either end of file, or end of buffer.
             bg_out.data_added_event.wait(1)
             bg_out.data_added_event.clear()
         # at this point it's still ambiguous if we are done reading or just full buffer.
         # Either way, if error (returned by ended process, or implied based on
         # presence of stuff in stderr output) we error out.
         # Else, we are happy.
         _returncode = _p.poll()
         if _returncode or (_returncode == None and bg_err.length):
             try:
                 _p.terminate()
             except:
                 pass
             bg_out.stop()
             bg_err.stop()
             err = '%s' % ''.join(bg_err)
             if err:
                 raise EnvironmentError("Subprocess exited due to an error:\n" + err)
             raise EnvironmentError("Subprocess exited with non 0 ret code:%s" % _returncode)
         self.process = _p
         self.output = bg_out
         self.error = bg_err
     def __iter__(self):
         return self
     def next(self):
         if self.process.poll():
             err = '%s' % ''.join(self.error)
             raise EnvironmentError("Subprocess exited due to an error:\n" + err)
         return self.output.next()
     def throw(self, type, value=None, traceback=None):
         if self.output.length or not self.output.done_reading:
             raise type(value)
     def close(self):
         try:
             self.process.terminate()
         except:
             pass
         try:
             self.output.close()
         except:
             pass
         try:
             self.error.close()
         except:
             pass
     def __del__(self):
         self.close()

rhodecode/lib/vcs/utils/__init__.py

➞

Show inline comments

 """
 This module provides some useful tools for ``vcs`` like annotate/diff html
 output. It also includes some internal helpers.
 """
 import sys
 import time
 import datetime
 def makedate():
     lt = time.localtime()
     if lt[8] == 1 and time.daylight:
         tz = time.altzone
     else:
         tz = time.timezone
     return time.mktime(lt), tz
 def aslist(obj, sep=None, strip=True):
     """
     Returns given string separated by sep as list
     :param obj:
     :param sep:
     :param strip:
     """
     if isinstance(obj, (basestring)):
         lst = obj.split(sep)
         if strip:
             lst = [v.strip() for v in lst]
         return lst
     elif isinstance(obj, (list, tuple)):
         return obj
     elif obj is None:
         return []
     else:
         return [obj]
 def date_fromtimestamp(unixts, tzoffset=0):
     """
     Makes a local datetime object out of unix timestamp
     :param unixts:
     :param tzoffset:
     """
     return datetime.datetime.fromtimestamp(float(unixts))
 def safe_int(val, default=None):
     """
     Returns int() of val if val is not convertable to int use default
     instead
     :param val:
     :param default:
     """
     try:
         val = int(val)
     except (ValueError, TypeError):
         val = default
     return val
 def safe_unicode(str_, from_encoding=None):
     """
     safe unicode function. Does few trick to turn str_ into unicode
     In case of UnicodeDecode error we try to return it with encoding detected
     by chardet library if it fails fallback to unicode with errors replaced
     :param str_: string to decode
     :rtype: unicode
     :returns: unicode object
     """
     from rhodecode.lib.utils2 import safe_unicode
     return safe_unicode(str_, from_encoding)
     if isinstance(str_, unicode):
         return str_
     if not from_encoding:
         from rhodecode.lib.vcs.conf import settings
         from_encoding = settings.DEFAULT_ENCODINGS
     if not isinstance(from_encoding, (list, tuple)):
         from_encoding = [from_encoding]
     try:
         return unicode(str_)
     except UnicodeDecodeError:
         pass
     try:
         return unicode(str_, from_encoding)
     except UnicodeDecodeError:
         pass
     for enc in from_encoding:
         try:
             return unicode(str_, enc)
         except UnicodeDecodeError:
             pass
     try:
         import chardet
         encoding = chardet.detect(str_)['encoding']
         if encoding is None:
             raise Exception()
         return str_.decode(encoding)
     except (ImportError, UnicodeDecodeError, Exception):
         return unicode(str_, from_encoding, 'replace')
+        return unicode(str_, from_encoding[0], 'replace')
 def safe_str(unicode_, to_encoding=None):
     """
     safe str function. Does few trick to turn unicode_ into string
     In case of UnicodeEncodeError we try to return it with encoding detected
     by chardet library if it fails fallback to string with errors replaced
     :param unicode_: unicode to encode
     :rtype: str
     :returns: str object
     """
     from rhodecode.lib.utils2 import safe_str
     return safe_str(unicode_, to_encoding)
     # if it's not basestr cast to str
     if not isinstance(unicode_, basestring):
         return str(unicode_)
     if isinstance(unicode_, str):
         return unicode_
     try:
         return unicode_.encode(to_encoding)
     except UnicodeEncodeError:
         pass
     if not to_encoding:
         from rhodecode.lib.vcs.conf import settings
         to_encoding = settings.DEFAULT_ENCODINGS
     if not isinstance(to_encoding, (list, tuple)):
         to_encoding = [to_encoding]
     for enc in to_encoding:
         try:
             return unicode_.encode(enc)
         except UnicodeEncodeError:
             pass
     try:
         import chardet
         encoding = chardet.detect(unicode_)['encoding']
         if encoding is None:
             raise UnicodeEncodeError()
         return unicode_.encode(encoding)
     except (ImportError, UnicodeEncodeError):
         return unicode_.encode(to_encoding, 'replace')
+        return unicode_.encode(to_encoding[0], 'replace')
     return safe_str
 def author_email(author):
     """
     returns email address of given author.
     If any of <,> sign are found, it fallbacks to regex findall()
     and returns first found result or empty string
     Regex taken from http://www.regular-expressions.info/email.html
     """
     import re
     r = author.find('>')
     l = author.find('<')
     if l == -1 or r == -1:
         # fallback to regex match of email out of a string
         email_re = re.compile(r"""[a-z0-9!#$%&'*+/=?^_`{|}~-]+(?:\.[a-z0-9!"""
                               r"""#$%&'*+/=?^_`{|}~-]+)*@(?:[a-z0-9](?:[a-z"""
                               r"""0-9-]*[a-z0-9])?\.)+[a-z0-9](?:[a-z0-9-]"""
                               r"""*[a-z0-9])?""", re.IGNORECASE)
         m = re.findall(email_re, author)
         return m[0] if m else ''
     return author[l + 1:r].strip()
 def author_name(author):
     """
     get name of author, or else username.
     It'll try to find an email in the author string and just cut it off
     to get the username
     """
     if not '@' in author:
         return author
     else:
         return author.replace(author_email(author), '').replace('<', '')\
             .replace('>', '').strip()

Changeset was too big and was cut off... Show full diff anyway

0 comments (0 inline, 0 general)