diff --git a/git/__init__.py b/git/__init__.py index 9ea811123..d87dcbdbb 100644 --- a/git/__init__.py +++ b/git/__init__.py @@ -15,13 +15,13 @@ def _init_externals(): """Initialize external projects by putting them into the path""" sys.path.append(os.path.join(os.path.dirname(__file__), 'ext', 'gitdb')) - + try: import gitdb except ImportError: raise ImportError("'gitdb' could not be found in your PYTHONPATH") #END verify import - + #} END initialization ################# @@ -41,14 +41,13 @@ def _init_externals(): from git.remote import * from git.index import * from git.util import ( - LockFile, - BlockingLockFile, + LockFile, + BlockingLockFile, Stats, Actor ) #} END imports -__all__ = [ name for name, obj in locals().items() - if not (name.startswith('_') or inspect.ismodule(obj)) ] - +__all__ = [name for name, obj in locals().items() + if not (name.startswith('_') or inspect.ismodule(obj))] diff --git a/git/cmd.py b/git/cmd.py index a1780de7e..042a528dc 100644 --- a/git/cmd.py +++ b/git/cmd.py @@ -6,73 +6,75 @@ import os, sys from util import ( - LazyMixin, + LazyMixin, stream_copy ) from exc import GitCommandError from subprocess import ( - call, + call, Popen, PIPE ) execute_kwargs = ('istream', 'with_keep_cwd', 'with_extended_output', - 'with_exceptions', 'as_process', - 'output_stream' ) + 'with_exceptions', 'as_process', + 'output_stream') __all__ = ('Git', ) + def dashify(string): return string.replace('_', '-') class Git(LazyMixin): + """ The Git class manages communication with the Git binary. - + It provides a convenient interface to calling the Git binary, such as in:: - + g = Git( git_dir ) g.init() # calls 'git init' program rval = g.ls_files() # calls 'git ls-files' program - + ``Debugging`` - Set the GIT_PYTHON_TRACE environment variable print each invocation + Set the GIT_PYTHON_TRACE environment variable print each invocation of the command to stdout. Set its value to 'full' to see details about the returned values. """ __slots__ = ("_working_dir", "cat_file_all", "cat_file_header", "_version_info", "_git_options") - + # CONFIGURATION # The size in bytes read from stdout when copying git's output to another stream - max_chunk_size = 1024*64 - + max_chunk_size = 1024 * 64 + git_exec_name = "git" # default that should work on linux and windows git_exec_name_win = "git.cmd" # alternate command name, windows only - + # Enables debugging of GitPython's git commands GIT_PYTHON_TRACE = os.environ.get("GIT_PYTHON_TRACE", False) - + # Provide the full path to the git executable. Otherwise it assumes git is in the path _git_exec_env_var = "GIT_PYTHON_GIT_EXECUTABLE" GIT_PYTHON_GIT_EXECUTABLE = os.environ.get(_git_exec_env_var, git_exec_name) - - + class AutoInterrupt(object): - """Kill/Interrupt the stored process instance once this instance goes out of scope. It is + + """Kill/Interrupt the stored process instance once this instance goes out of scope. It is used to prevent processes piling up in case iterators stop reading. Besides all attributes are wired through to the contained process object. - + The wait method was overridden to perform automatic status code checking and possibly raise.""" - __slots__= ("proc", "args") - - def __init__(self, proc, args ): + __slots__ = ("proc", "args") + + def __init__(self, proc, args): self.proc = proc self.args = args - + def __del__(self): self.proc.stdout.close() self.proc.stderr.close() @@ -80,60 +82,61 @@ def __del__(self): # did the process finish already so we have a return code ? if self.proc.poll() is not None: return - - # can be that nothing really exists anymore ... + + # can be that nothing really exists anymore ... if os is None: return - + # try to kill it try: os.kill(self.proc.pid, 2) # interrupt signal self.proc.wait() # ensure process goes away except OSError: - pass # ignore error when process already died + pass # ignore error when process already died except AttributeError: - # try windows - # for some reason, providing None for stdout/stderr still prints something. This is why - # we simply use the shell and redirect to nul. Its slower than CreateProcess, question + # try windows + # for some reason, providing None for stdout/stderr still prints something. This is why + # we simply use the shell and redirect to nul. Its slower than CreateProcess, question # is whether we really want to see all these messages. Its annoying no matter what. call(("TASKKILL /F /T /PID %s 2>nul 1>nul" % str(self.proc.pid)), shell=True) - # END exception handling - + # END exception handling + def __getattr__(self, attr): return getattr(self.proc, attr) - + def wait(self): - """Wait for the process and return its status code. - + """Wait for the process and return its status code. + :raise GitCommandError: if the return status is not 0""" status = self.proc.wait() if status != 0: raise GitCommandError(self.args, status, self.proc.stderr.read()) - # END status handling + # END status handling return status # END auto interrupt - + class CatFileContentStream(object): - """Object representing a sized read-only stream returning the contents of + + """Object representing a sized read-only stream returning the contents of an object. - It behaves like a stream, but counts the data read and simulates an empty + It behaves like a stream, but counts the data read and simulates an empty stream once our sized content region is empty. - If not all data is read to the end of the objects's lifetime, we read the + If not all data is read to the end of the objects's lifetime, we read the rest to assure the underlying stream continues to work""" - + __slots__ = ('_stream', '_nbr', '_size') - + def __init__(self, size, stream): self._stream = stream self._size = size self._nbr = 0 # num bytes read - - # special case: if the object is empty, has null bytes, get the + + # special case: if the object is empty, has null bytes, get the # final newline right away. if size == 0: stream.read(1) # END handle empty streams - + def read(self, size=-1): bytes_left = self._size - self._nbr if bytes_left == 0: @@ -147,17 +150,17 @@ def read(self, size=-1): # END check early depletion data = self._stream.read(size) self._nbr += len(data) - + # check for depletion, read our final byte to make the stream usable by others if self._size - self._nbr == 0: self._stream.read(1) # final newline # END finish reading return data - + def readline(self, size=-1): if self._nbr == self._size: return '' - + # clamp size to lowest allowed value bytes_left = self._size - self._nbr if size > -1: @@ -165,21 +168,21 @@ def readline(self, size=-1): else: size = bytes_left # END handle size - + data = self._stream.readline(size) self._nbr += len(data) - + # handle final byte if self._size - self._nbr == 0: self._stream.read(1) # END finish reading - + return data - + def readlines(self, size=-1): if self._nbr == self._size: return list() - + # leave all additional logic to our readline method, we just check the size out = list() nbr = 0 @@ -195,16 +198,16 @@ def readlines(self, size=-1): # END handle size constraint # END readline loop return out - + def __iter__(self): return self - + def next(self): line = self.readline() if not line: raise StopIteration return line - + def __del__(self): bytes_left = self._size - self._nbr if bytes_left: @@ -212,15 +215,14 @@ def __del__(self): # includes terminating newline self._stream.read(bytes_left + 1) # END handle incomplete read - - + def __init__(self, working_dir=None): """Initialize this instance with: - + :param working_dir: - Git directory we should work in. If None, we always work in the current + Git directory we should work in. If None, we always work in the current directory as returned by os.getcwd(). - It is meant to be the working tree directory if available, or the + It is meant to be the working tree directory if available, or the .git directory in case of bare repositories.""" super(Git, self).__init__() self._working_dir = working_dir @@ -231,7 +233,7 @@ def __init__(self, working_dir=None): self.cat_file_all = None def __getattr__(self, name): - """A convenience method as it allows to call the command as if it was + """A convenience method as it allows to call the command as if it was an object. :return: Callable object that will execute call _call_process with your arguments.""" if name[0] == '_': @@ -246,13 +248,12 @@ def _set_cache_(self, attr): else: super(Git, self)._set_cache_(attr) #END handle version info - @property def working_dir(self): """:return: Git directory we are working on""" return self._working_dir - + @property def version_info(self): """ @@ -266,8 +267,8 @@ def execute(self, command, with_keep_cwd=False, with_extended_output=False, with_exceptions=True, - as_process=False, - output_stream=None, + as_process=False, + output_stream=None, **subprocess_kwargs ): """Handles executing the command on the shell and consumes and returns @@ -293,43 +294,43 @@ def execute(self, command, Whether to raise an exception when git returns a non-zero status. :param as_process: - Whether to return the created process instance directly from which - streams can be read on demand. This will render with_extended_output and - with_exceptions ineffective - the caller will have + Whether to return the created process instance directly from which + streams can be read on demand. This will render with_extended_output and + with_exceptions ineffective - the caller will have to deal with the details himself. It is important to note that the process will be placed into an AutoInterrupt - wrapper that will interrupt the process once it goes out of scope. If you - use the command in iterators, you should pass the whole process instance + wrapper that will interrupt the process once it goes out of scope. If you + use the command in iterators, you should pass the whole process instance instead of a single stream. - + :param output_stream: - If set to a file-like object, data produced by the git command will be + If set to a file-like object, data produced by the git command will be output to the given stream directly. This feature only has any effect if as_process is False. Processes will always be created with a pipe due to issues with subprocess. - This merely is a workaround as data will be copied from the + This merely is a workaround as data will be copied from the output pipe to the given output stream directly. - + :param subprocess_kwargs: - Keyword arguments to be passed to subprocess.Popen. Please note that - some of the valid kwargs are already set by this method, the ones you + Keyword arguments to be passed to subprocess.Popen. Please note that + some of the valid kwargs are already set by this method, the ones you specify may not be the same ones. - + :return: * str(output) if extended_output = False (Default) * tuple(int(status), str(stdout), str(stderr)) if extended_output = True - + if ouput_stream is True, the stdout value will be your output stream: * output_stream if extended_output = False * tuple(int(status), output_stream, str(stderr)) if extended_output = True Note git is executed with LC_MESSAGES="C" to ensure consitent output regardless of system language. - + :raise GitCommandError: - + :note: - If you add additional keyword arguments to the signature of this method, + If you add additional keyword arguments to the signature of this method, you must update the execute_kwargs tuple housed in this module.""" if self.GIT_PYTHON_TRACE and not self.GIT_PYTHON_TRACE == 'full': print ' '.join(command) @@ -338,8 +339,8 @@ def execute(self, command, if with_keep_cwd or self._working_dir is None: cwd = os.getcwd() else: - cwd=self._working_dir - + cwd = self._working_dir + # Start the process proc = Popen(command, env={"LC_MESSAGES": "C"}, @@ -347,19 +348,19 @@ def execute(self, command, stdin=istream, stderr=PIPE, stdout=PIPE, - close_fds=(os.name=='posix'),# unsupported on linux + close_fds=(os.name == 'posix'), # unsupported on linux **subprocess_kwargs ) if as_process: return self.AutoInterrupt(proc, command) - + # Wait for the process to return status = 0 stdout_value = '' stderr_value = '' try: if output_stream is None: - stdout_value, stderr_value = proc.communicate() + stdout_value, stderr_value = proc.communicate() # strip trailing "\n" if stdout_value.endswith("\n"): stdout_value = stdout_value[:-1] @@ -422,18 +423,18 @@ def transform_kwargs(self, split_single_char_options=False, **kwargs): @classmethod def __unpack_args(cls, arg_list): - if not isinstance(arg_list, (list,tuple)): + if not isinstance(arg_list, (list, tuple)): if isinstance(arg_list, unicode): return [arg_list.encode('utf-8')] - return [ str(arg_list) ] - + return [str(arg_list)] + outlist = list() for arg in arg_list: if isinstance(arg_list, (list, tuple)): - outlist.extend(cls.__unpack_args( arg )) + outlist.extend(cls.__unpack_args(arg)) elif isinstance(arg_list, unicode): outlist.append(arg_list.encode('utf-8')) - # END recursion + # END recursion else: outlist.append(str(arg)) # END for each arg @@ -488,10 +489,10 @@ def _call_process(self, method, *args, **kwargs): # Prepare the argument list opt_args = self.transform_kwargs(**kwargs) - + ext_args = self.__unpack_args([a for a in args if a is not None]) args = opt_args + ext_args - + def make_call(): call = [self.GIT_PYTHON_GIT_EXECUTABLE] @@ -504,7 +505,7 @@ def make_call(): call.extend(args) return call #END utility to recreate call after changes - + if sys.platform == 'win32': try: try: @@ -516,13 +517,13 @@ def make_call(): #END handle overridden variable type(self).GIT_PYTHON_GIT_EXECUTABLE = self.git_exec_name_win call = [self.GIT_PYTHON_GIT_EXECUTABLE] + list(args) - + try: return self.execute(make_call(), **_kwargs) finally: import warnings msg = "WARNING: Automatically switched to use git.cmd as git executable, which reduces performance by ~70%." - msg += "Its recommended to put git.exe into the PATH or to set the %s environment variable to the executable's location" % self._git_exec_env_var + msg += "Its recommended to put git.exe into the PATH or to set the %s environment variable to the executable's location" % self._git_exec_env_var warnings.warn(msg) #END print of warning #END catch first failure @@ -532,15 +533,15 @@ def make_call(): else: return self.execute(make_call(), **_kwargs) #END handle windows default installation - + def _parse_object_header(self, header_line): """ :param header_line: type_string size_as_int - + :return: (hex_sha, type_string, size_as_int) - - :raise ValueError: if the header contains indication for an error due to + + :raise ValueError: if the header contains indication for an error due to incorrect input sha""" tokens = header_line.split() if len(tokens) != 3: @@ -550,46 +551,46 @@ def _parse_object_header(self, header_line): raise ValueError("SHA %s could not be resolved, git returned: %r" % (tokens[0], header_line.strip())) # END handle actual return value # END error handling - + if len(tokens[0]) != 40: - raise ValueError("Failed to parse header: %r" % header_line) + raise ValueError("Failed to parse header: %r" % header_line) return (tokens[0], tokens[1], int(tokens[2])) - + def __prepare_ref(self, ref): # required for command to separate refs on stdin refstr = str(ref) # could be ref-object if refstr.endswith("\n"): return refstr return refstr + "\n" - - def __get_persistent_cmd(self, attr_name, cmd_name, *args,**kwargs): + + def __get_persistent_cmd(self, attr_name, cmd_name, *args, **kwargs): cur_val = getattr(self, attr_name) if cur_val is not None: return cur_val - - options = { "istream" : PIPE, "as_process" : True } - options.update( kwargs ) - - cmd = self._call_process( cmd_name, *args, **options ) - setattr(self, attr_name, cmd ) + + options = {"istream": PIPE, "as_process": True} + options.update(kwargs) + + cmd = self._call_process(cmd_name, *args, **options) + setattr(self, attr_name, cmd) return cmd - + def __get_object_header(self, cmd, ref): cmd.stdin.write(self.__prepare_ref(ref)) cmd.stdin.flush() return self._parse_object_header(cmd.stdout.readline()) - + def get_object_header(self, ref): - """ Use this method to quickly examine the type and size of the object behind - the given ref. - - :note: The method will only suffer from the costs of command invocation - once and reuses the command in subsequent calls. - + """ Use this method to quickly examine the type and size of the object behind + the given ref. + + :note: The method will only suffer from the costs of command invocation + once and reuses the command in subsequent calls. + :return: (hexsha, type_string, size_as_int)""" cmd = self.__get_persistent_cmd("cat_file_header", "cat_file", batch_check=True) return self.__get_object_header(cmd, ref) - + def get_object_data(self, ref): """ As get_object_header, but returns object data as well :return: (hexsha, type_string, size_as_int,data_string) @@ -598,7 +599,7 @@ def get_object_data(self, ref): data = stream.read(size) del(stream) return (hexsha, typename, size, data) - + def stream_object_data(self, ref): """As get_object_header, but returns the data as a stream :return: (hexsha, type_string, size_as_int, stream) @@ -607,12 +608,12 @@ def stream_object_data(self, ref): cmd = self.__get_persistent_cmd("cat_file_all", "cat_file", batch=True) hexsha, typename, size = self.__get_object_header(cmd, ref) return (hexsha, typename, size, self.CatFileContentStream(size, cmd.stdout)) - + def clear_cache(self): """Clear all kinds of internal caches to release resources. - + Currently persistent commands will be interrupted. - + :return: self""" self.cat_file_all = None self.cat_file_header = None diff --git a/git/config.py b/git/config.py index 285ade6b7..15aa76f0d 100644 --- a/git/config.py +++ b/git/config.py @@ -17,7 +17,9 @@ __all__ = ('GitConfigParser', 'SectionConstraint') + class MetaParserBuilder(type): + """Utlity class wrapping base-class methods into decorators that assure read-only properties""" def __new__(metacls, name, bases, clsdict): """ @@ -27,7 +29,7 @@ def __new__(metacls, name, bases, clsdict): if kmm in clsdict: mutating_methods = clsdict[kmm] for base in bases: - methods = ( t for t in inspect.getmembers(base, inspect.ismethod) if not t[0].startswith("_") ) + methods = (t for t in inspect.getmembers(base, inspect.ismethod) if not t[0].startswith("_")) for name, method in methods: if name in clsdict: continue @@ -35,30 +37,32 @@ def __new__(metacls, name, bases, clsdict): if name in mutating_methods: method_with_values = set_dirty_and_flush_changes(method_with_values) # END mutating methods handling - + clsdict[name] = method_with_values # END for each name/method pair # END for each base # END if mutating methods configuration is set - + new_type = super(MetaParserBuilder, metacls).__new__(metacls, name, bases, clsdict) return new_type - - + def needs_values(func): """Returns method assuring we read values (on demand) before we try to access them""" + def assure_data_present(self, *args, **kwargs): self.read() return func(self, *args, **kwargs) # END wrapper method assure_data_present.__name__ = func.__name__ return assure_data_present - + + def set_dirty_and_flush_changes(non_const_func): """Return method that checks whether given non constant function may be called. If so, the instance will be set dirty. Additionally, we flush the changes right to disk""" + def flush_changes(self, *args, **kwargs): rval = non_const_func(self, *args, **kwargs) self.write() @@ -66,64 +70,65 @@ def flush_changes(self, *args, **kwargs): # END wrapper method flush_changes.__name__ = non_const_func.__name__ return flush_changes - + class SectionConstraint(object): - """Constrains a ConfigParser to only option commands which are constrained to + + """Constrains a ConfigParser to only option commands which are constrained to always use the section we have been initialized with. - + It supports all ConfigParser methods that operate on an option""" __slots__ = ("_config", "_section_name") - _valid_attrs_ = ("get_value", "set_value", "get", "set", "getint", "getfloat", "getboolean", "has_option", + _valid_attrs_ = ("get_value", "set_value", "get", "set", "getint", "getfloat", "getboolean", "has_option", "remove_section", "remove_option", "options") - + def __init__(self, config, section): self._config = config self._section_name = section - + def __getattr__(self, attr): if attr in self._valid_attrs_: return lambda *args, **kwargs: self._call_config(attr, *args, **kwargs) - return super(SectionConstraint,self).__getattribute__(attr) - + return super(SectionConstraint, self).__getattribute__(attr) + def _call_config(self, method, *args, **kwargs): - """Call the configuration at the given method which must take a section name + """Call the configuration at the given method which must take a section name as first argument""" return getattr(self._config, method)(self._section_name, *args, **kwargs) - + @property def config(self): """return: Configparser instance we constrain""" return self._config - + class GitConfigParser(cp.RawConfigParser, object): + """Implements specifics required to read git style configuration files. - + This variation behaves much like the git.config command such that the configuration will be read on demand based on the filepath given during initialization. - - The changes will automatically be written once the instance goes out of scope, but + + The changes will automatically be written once the instance goes out of scope, but can be triggered manually as well. - - The configuration file will be locked if you intend to change values preventing other + + The configuration file will be locked if you intend to change values preventing other instances to write concurrently. - + :note: The config is case-sensitive even when queried, hence section and option names must match perfectly.""" __metaclass__ = MetaParserBuilder - - + #{ Configuration # The lock type determines the type of lock to use in new configuration readers. # They must be compatible to the LockFile interface. # A suitable alternative would be the BlockingLockFile t_lock = LockFile re_comment = re.compile('^\s*[#;]') - - #} END configuration - + + #} END configuration + OPTCRE = re.compile( r'\s*(?P