René's URL Explorer Experiment


Title: No title

direct link

Domain: github.com


Hey, it has address:
``. One of these helpers (``ext::``) can be used to invoke any arbitrary command. See: - https://git-scm.com/docs/gitremote-helpers - https://git-scm.com/docs/git-remote-ext """ match = cls.re_unsafe_protocol.match(url) if match: protocol = match.group(1) raise UnsafeProtocolError( f"The `{protocol}::` protocol looks suspicious, use `allow_unsafe_protocols=True` to allow it." ) @classmethod def _canonicalize_option_name(cls, option: str) -> str: """Return the option name used for unsafe-option checks. Examples: ``"--upload-pack=/tmp/helper"`` -> ``"upload-pack"`` ``"upload_pack"`` -> ``"upload-pack"`` ``"--config core.filemode=false"`` -> ``"config"`` """ option_name = option.lstrip("-").split("=", 1)[0] option_tokens = option_name.split(None, 1) if not option_tokens: return "" return dashify(option_tokens[0]) @classmethod def check_unsafe_options(cls, options: List[str], unsafe_options: List[str]) -> None: """Check for unsafe options. Some options that are passed to ``git `` can be used to execute arbitrary commands. These are blocked by default. """ # Options can be of the form `foo`, `--foo`, `--foo bar`, or `--foo=bar`. canonical_unsafe_options = {cls._canonicalize_option_name(option): option for option in unsafe_options} for option in options: unsafe_option = canonical_unsafe_options.get(cls._canonicalize_option_name(option)) if unsafe_option is not None: raise UnsafeOptionError(f"{unsafe_option} is not allowed, use `allow_unsafe_options=True` to allow it.") AutoInterrupt: TypeAlias = _AutoInterrupt CatFileContentStream: TypeAlias = _CatFileContentStream def __init__(self, working_dir: Union[None, PathLike] = None) -> None: """Initialize this instance with: :param working_dir: Git directory we should work in. If ``None``, we always work in the current directory as returned by :func:`os.getcwd`. This is meant to be the working tree directory if available, or the ``.git`` directory in case of bare repositories. """ super().__init__() self._working_dir = expand_path(working_dir) self._git_options: Union[List[str], Tuple[str, ...]] = () self._persistent_git_options: List[str] = [] # Extra environment variables to pass to git commands self._environment: Dict[str, str] = {} # Cached version slots self._version_info: Union[Tuple[int, ...], None] = None self._version_info_token: object = None # Cached command slots self.cat_file_header: Union[None, TBD] = None self.cat_file_all: Union[None, TBD] = None def __getattribute__(self, name: str) -> Any: if name == "USE_SHELL": _warn_use_shell(extra_danger=False) return super().__getattribute__(name) def __getattr__(self, name: str) -> Any: """A convenience method as it allows to call the command as if it was an object. :return: Callable object that will execute call :meth:`_call_process` with your arguments. """ if name.startswith("_"): return super().__getattribute__(name) return lambda *args, **kwargs: self._call_process(name, *args, **kwargs) def set_persistent_git_options(self, **kwargs: Any) -> None: """Specify command line options to the git executable for subsequent subcommand calls. :param kwargs: A dict of keyword arguments. These arguments are passed as in :meth:`_call_process`, but will be passed to the git command rather than the subcommand. """ self._persistent_git_options = self.transform_kwargs(split_single_char_options=True, **kwargs) @property def working_dir(self) -> Union[None, PathLike]: """:return: Git directory we are working on""" return self._working_dir @property def version_info(self) -> Tuple[int, ...]: """ :return: Tuple with integers representing the major, minor and additional version numbers as parsed from :manpage:`git-version(1)`. Up to four fields are used. This value is generated on demand and is cached. """ # Refreshing is global, but version_info caching is per-instance. refresh_token = self._refresh_token # Copy token in case of concurrent refresh. # Use the cached version if obtained after the most recent refresh. if self._version_info_token is refresh_token: assert self._version_info is not None, "Bug: corrupted token-check state" return self._version_info # Run "git version" and parse it. process_version = self._call_process("version") version_string = process_version.split(" ")[2] version_fields = version_string.split(".")[:4] leading_numeric_fields = itertools.takewhile(str.isdigit, version_fields) self._version_info = tuple(map(int, leading_numeric_fields)) # This value will be considered valid until the next refresh. self._version_info_token = refresh_token return self._version_info @overload def execute( self, command: Union[str, Sequence[Any]], *, as_process: Literal[True], ) -> "AutoInterrupt": ... @overload def execute( self, command: Union[str, Sequence[Any]], *, as_process: Literal[False] = False, stdout_as_string: Literal[True], ) -> Union[str, Tuple[int, str, str]]: ... @overload def execute( self, command: Union[str, Sequence[Any]], *, as_process: Literal[False] = False, stdout_as_string: Literal[False] = False, ) -> Union[bytes, Tuple[int, bytes, str]]: ... @overload def execute( self, command: Union[str, Sequence[Any]], *, with_extended_output: Literal[False], as_process: Literal[False], stdout_as_string: Literal[True], ) -> str: ... @overload def execute( self, command: Union[str, Sequence[Any]], *, with_extended_output: Literal[False], as_process: Literal[False], stdout_as_string: Literal[False], ) -> bytes: ... def execute( self, command: Union[str, Sequence[Any]], istream: Union[None, BinaryIO] = None, with_extended_output: bool = False, with_exceptions: bool = True, as_process: bool = False, output_stream: Union[None, BinaryIO] = None, stdout_as_string: bool = True, kill_after_timeout: Union[None, float] = None, with_stdout: bool = True, universal_newlines: bool = False, shell: Union[None, bool] = None, env: Union[None, Mapping[str, str]] = None, max_chunk_size: int = io.DEFAULT_BUFFER_SIZE, strip_newline_in_stdout: bool = True, **subprocess_kwargs: Any, ) -> Union[str, bytes, Tuple[int, Union[str, bytes], str], AutoInterrupt]: R"""Handle executing the command, and consume and return the returned information (stdout). :param command: The command argument list to execute. It should be a sequence of program arguments, or a string. The program to execute is the first item in the args sequence or string. :param istream: Standard input filehandle passed to :class:`subprocess.Popen`. :param with_extended_output: Whether to return a (status, stdout, stderr) tuple. :param with_exceptions: 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 to deal with the details. It is important to note that the process will be placed into an :class:`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 instead of a single stream. :param output_stream: If set to a file-like object, data produced by the git command will be copied to the given stream instead of being returned as a string. This feature only has any effect if `as_process` is ``False``. :param stdout_as_string: If ``False``, the command's standard output will be bytes. Otherwise, it will be decoded into a string using the default encoding (usually UTF-8). The latter can fail, if the output contains binary data. :param kill_after_timeout: Specifies a timeout in seconds for the git command, after which the process should be killed. This will have no effect if `as_process` is set to ``True``. It is set to ``None`` by default and will let the process run until the timeout is explicitly specified. Uses of this feature should be carefully considered, due to the following limitations: 1. This feature is not supported at all on Windows. 2. Effectiveness may vary by operating system. ``ps --ppid`` is used to enumerate child processes, which is available on most GNU/Linux systems but not most others. 3. Deeper descendants do not receive signals, though they may sometimes terminate as a consequence of their parent processes being killed. 4. `kill_after_timeout` uses ``SIGKILL``, which can have negative side effects on a repository. For example, stale locks in case of :manpage:`git-gc(1)` could render the repository incapable of accepting changes until the lock is manually removed. :param with_stdout: If ``True``, default ``True``, we open stdout on the created process. :param universal_newlines: If ``True``, pipes will be opened as text, and lines are split at all known line endings. :param shell: Whether to invoke commands through a shell (see :class:`Popen(..., shell=True) `). If this is not ``None``, it overrides :attr:`USE_SHELL`. Passing ``shell=True`` to this or any other GitPython function should be avoided, as it is unsafe under most circumstances. This is because it is typically not feasible to fully consider and account for the effect of shell expansions, especially when passing ``shell=True`` to other methods that forward it to :meth:`Git.execute`. Passing ``shell=True`` is also no longer needed (nor useful) to work around any known operating system specific issues. :param env: A dictionary of environment variables to be passed to :class:`subprocess.Popen`. :param max_chunk_size: Maximum number of bytes in one chunk of data passed to the `output_stream` in one invocation of its ``write()`` method. If the given number is not positive then the default value is used. :param strip_newline_in_stdout: Whether to strip the trailing ``\n`` of the command stdout. :param subprocess_kwargs: Keyword arguments to be passed to :class:`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` is ``False`` (Default) * tuple(int(status), str(stdout), str(stderr)), if `extended_output` is ``True`` If `output_stream` is ``True``, the stdout value will be your output stream: * output_stream, if `extended_output` is ``False`` * tuple(int(status), output_stream, str(stderr)), if `extended_output` is ``True`` Note that git is executed with ``LC_MESSAGES="C"`` to ensure consistent output regardless of system language. :raise git.exc.GitCommandError: :note: If you add additional keyword arguments to the signature of this method, you must update the ``execute_kwargs`` variable housed in this module. """ # Remove password for the command if present. redacted_command = remove_password_if_present(command) if self.GIT_PYTHON_TRACE and (self.GIT_PYTHON_TRACE != "full" or as_process): _logger.info(" ".join(redacted_command)) # Allow the user to have the command executed in their working dir. try: cwd = self._working_dir or os.getcwd() # type: Union[None, str] if not os.access(str(cwd), os.X_OK): cwd = None except FileNotFoundError: cwd = None # Start the process. inline_env = env env = os.environ.copy() # Attempt to force all output to plain ASCII English, which is what some parsing # code may expect. # According to https://askubuntu.com/a/311796, we are setting LANGUAGE as well # just to be sure. env["LANGUAGE"] = "C" env["LC_ALL"] = "C" env.update(self._environment) if inline_env is not None: env.update(inline_env) if sys.platform == "win32": if kill_after_timeout is not None: raise GitCommandError( redacted_command, '"kill_after_timeout" feature is not supported on Windows.', ) cmd_not_found_exception = OSError else: cmd_not_found_exception = FileNotFoundError # END handle stdout_sink = PIPE if with_stdout else getattr(subprocess, "DEVNULL", None) or open(os.devnull, "wb") if shell is None: # Get the value of USE_SHELL with no deprecation warning. Do this without # warnings.catch_warnings, to avoid a race condition with application code # configuring warnings. The value could be looked up in type(self).__dict__ # or Git.__dict__, but those can break under some circumstances. This works # the same as self.USE_SHELL in more situations; see Git.__getattribute__. shell = super().__getattribute__("USE_SHELL") _logger.debug( "Popen(%s, cwd=%s, stdin=%s, shell=%s, universal_newlines=%s)", redacted_command, cwd, "" if istream else "None", shell, universal_newlines, ) try: proc = safer_popen( command, env=env, cwd=cwd, bufsize=-1, stdin=(istream or DEVNULL), stderr=PIPE, stdout=stdout_sink, shell=shell, universal_newlines=universal_newlines, encoding=defenc if universal_newlines else None, **subprocess_kwargs, ) except cmd_not_found_exception as err: raise GitCommandNotFound(redacted_command, err) from err else: # Replace with a typeguard for Popen[bytes]? proc.stdout = cast(BinaryIO, proc.stdout) proc.stderr = cast(BinaryIO, proc.stderr) if as_process: return self.AutoInterrupt(proc, command) if sys.platform != "win32" and kill_after_timeout is not None: # Help mypy figure out this is not None even when used inside communicate(). timeout = kill_after_timeout def kill_process(pid: int) -> None: """Callback to kill a process. This callback implementation would be ineffective and unsafe on Windows. """ p = Popen(["ps", "--ppid", str(pid)], stdout=PIPE) child_pids = [] if p.stdout is not None: for line in p.stdout: if len(line.split()) > 0: local_pid = (line.split())[0] if local_pid.isdigit(): child_pids.append(int(local_pid)) try: os.kill(pid, signal.SIGKILL) for child_pid in child_pids: try: os.kill(child_pid, signal.SIGKILL) except OSError: pass # Tell the main routine that the process was killed. kill_check.set() except OSError: # It is possible that the process gets completed in the duration # after timeout happens and before we try to kill the process. pass return def communicate() -> Tuple[AnyStr, AnyStr]: watchdog.start() out, err = proc.communicate() watchdog.cancel() if kill_check.is_set(): err = 'Timeout: the command "%s" did not complete in %d secs.' % ( " ".join(redacted_command), timeout, ) if not universal_newlines: err = err.encode(defenc) return out, err # END helpers kill_check = threading.Event() watchdog = threading.Timer(timeout, kill_process, args=(proc.pid,)) else: communicate = proc.communicate # Wait for the process to return. status = 0 stdout_value: Union[str, bytes] = b"" stderr_value: Union[str, bytes] = b"" newline = "\n" if universal_newlines else b"\n" try: if output_stream is None: stdout_value, stderr_value = communicate() # Strip trailing "\n". if stdout_value is not None and stdout_value.endswith(newline) and strip_newline_in_stdout: # type: ignore[arg-type] stdout_value = stdout_value[:-1] if stderr_value is not None and stderr_value.endswith(newline): # type: ignore[arg-type] stderr_value = stderr_value[:-1] status = proc.returncode else: max_chunk_size = max_chunk_size if max_chunk_size and max_chunk_size > 0 else io.DEFAULT_BUFFER_SIZE if proc.stdout is not None: stream_copy(proc.stdout, output_stream, max_chunk_size) stdout_value = proc.stdout.read() if proc.stderr is not None: stderr_value = proc.stderr.read() # Strip trailing "\n". if stderr_value is not None and stderr_value.endswith(newline): # type: ignore[arg-type] stderr_value = stderr_value[:-1] status = proc.wait() # END stdout handling finally: if proc.stdout is not None: proc.stdout.close() if proc.stderr is not None: proc.stderr.close() if self.GIT_PYTHON_TRACE == "full": cmdstr = " ".join(redacted_command) def as_text(stdout_value: Union[bytes, str]) -> str: return not output_stream and safe_decode(stdout_value) or "" # END as_text if stderr_value: _logger.info( "%s -> %d; stdout: '%s'; stderr: '%s'", cmdstr, status, as_text(stdout_value), safe_decode(stderr_value), ) elif stdout_value: _logger.info("%s -> %d; stdout: '%s'", cmdstr, status, as_text(stdout_value)) else: _logger.info("%s -> %d", cmdstr, status) # END handle debug printing if with_exceptions and status != 0: raise GitCommandError(redacted_command, status, stderr_value, stdout_value) if isinstance(stdout_value, bytes) and stdout_as_string: # Could also be output_stream. stdout_value = safe_decode(stdout_value) # Allow access to the command's status code. if with_extended_output: return (status, stdout_value, safe_decode(stderr_value)) else: return stdout_value def environment(self) -> Dict[str, str]: return self._environment def update_environment(self, **kwargs: Any) -> Dict[str, Union[str, None]]: """Set environment variables for future git invocations. Return all changed values in a format that can be passed back into this function to revert the changes. Examples:: old_env = self.update_environment(PWD='/tmp') self.update_environment(**old_env) :param kwargs: Environment variables to use for git processes. :return: Dict that maps environment variables to their old values """ old_env = {} for key, value in kwargs.items(): # Set value if it is None. if value is not None: old_env[key] = self._environment.get(key) self._environment[key] = value # Remove key from environment if its value is None. elif key in self._environment: old_env[key] = self._environment[key] del self._environment[key] return old_env @contextlib.contextmanager def custom_environment(self, **kwargs: Any) -> Iterator[None]: """A context manager around the above :meth:`update_environment` method to restore the environment back to its previous state after operation. Examples:: with self.custom_environment(GIT_SSH='/bin/ssh_wrapper'): repo.remotes.origin.fetch() :param kwargs: See :meth:`update_environment`. """ old_env = self.update_environment(**kwargs) try: yield finally: self.update_environment(**old_env) def transform_kwarg(self, name: str, value: Any, split_single_char_options: bool) -> List[str]: if len(name) == 1: if value is True: return ["-%s" % name] elif value not in (False, None): if split_single_char_options: return ["-%s" % name, "%s" % value] else: return ["-%s%s" % (name, value)] else: if value is True: return ["--%s" % dashify(name)] elif value is not False and value is not None: return ["--%s=%s" % (dashify(name), value)] return [] def transform_kwargs(self, split_single_char_options: bool = True, **kwargs: Any) -> List[str]: """Transform Python-style kwargs into git command line options.""" args = [] for k, v in kwargs.items(): if isinstance(v, (list, tuple)): for value in v: args += self.transform_kwarg(k, value, split_single_char_options) else: args += self.transform_kwarg(k, v, split_single_char_options) return args @classmethod def _unpack_args(cls, arg_list: Sequence[str]) -> List[str]: outlist = [] if isinstance(arg_list, (list, tuple)): for arg in arg_list: outlist.extend(cls._unpack_args(arg)) else: outlist.append(str(arg_list)) return outlist def __call__(self, **kwargs: Any) -> "Git": """Specify command line options to the git executable for a subcommand call. :param kwargs: A dict of keyword arguments. These arguments are passed as in :meth:`_call_process`, but will be passed to the git command rather than the subcommand. Examples:: git(work_tree='/tmp').difftool() """ self._git_options = self.transform_kwargs(split_single_char_options=True, **kwargs) return self @overload def _call_process( self, method: str, *args: None, **kwargs: None ) -> str: ... # If no args were given, execute the call with all defaults. @overload def _call_process( self, method: str, istream: int, as_process: Literal[True], *args: Any, **kwargs: Any, ) -> "Git.AutoInterrupt": ... @overload def _call_process( self, method: str, *args: Any, **kwargs: Any ) -> Union[str, bytes, Tuple[int, Union[str, bytes], str], "Git.AutoInterrupt"]: ... def _call_process( self, method: str, *args: Any, **kwargs: Any ) -> Union[str, bytes, Tuple[int, Union[str, bytes], str], "Git.AutoInterrupt"]: """Run the given git command with the specified arguments and return the result as a string. :param method: The command. Contained ``_`` characters will be converted to hyphens, such as in ``ls_files`` to call ``ls-files``. :param args: The list of arguments. If ``None`` is included, it will be pruned. This allows your commands to call git more conveniently, as ``None`` is realized as non-existent. :param kwargs: Contains key-values for the following: - The :meth:`execute()` kwds, as listed in ``execute_kwargs``. - "Command options" to be converted by :meth:`transform_kwargs`. - The ``insert_kwargs_after`` key which its value must match one of ``*args``. It also contains any command options, to be appended after the matched arg. Examples:: git.rev_list('master', max_count=10, header=True) turns into:: git rev-list --max-count=10 --header=master :return: Same as :meth:`execute`. If no args are given, used :meth:`execute`'s default (especially ``as_process = False``, ``stdout_as_string = True``) and return :class:`str`. """ # Handle optional arguments prior to calling transform_kwargs. # Otherwise these'll end up in args, which is bad. exec_kwargs = {k: v for k, v in kwargs.items() if k in execute_kwargs} opts_kwargs = {k: v for k, v in kwargs.items() if k not in execute_kwargs} insert_after_this_arg = opts_kwargs.pop("insert_kwargs_after", None) # Prepare the argument list. opt_args = self.transform_kwargs(**opts_kwargs) ext_args = self._unpack_args([a for a in args if a is not None]) if insert_after_this_arg is None: args_list = opt_args + ext_args else: try: index = ext_args.index(insert_after_this_arg) except ValueError as err: raise ValueError( "Couldn't find argument '%s' in args %s to insert cmd options after" % (insert_after_this_arg, str(ext_args)) ) from err # END handle error args_list = ext_args[: index + 1] + opt_args + ext_args[index + 1 :] # END handle opts_kwargs call = [self.GIT_PYTHON_GIT_EXECUTABLE] # Add persistent git options. call.extend(self._persistent_git_options) # Add the git options, then reset to empty to avoid side effects. call.extend(self._git_options) self._git_options = () call.append(dashify(method)) call.extend(args_list) return self.execute(call, **exec_kwargs) def _parse_object_header(self, header_line: str) -> Tuple[str, str, int]: """ :param header_line: A line of the form:: 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 incorrect input sha. """ tokens = header_line.split() if len(tokens) != 3: if not tokens: err_msg = ( f"SHA is empty, possible dubious ownership in the repository " f"""at {self._working_dir}.\n If this is unintended run:\n\n """ f""" "git config --global --add safe.directory {self._working_dir}" """ ) raise ValueError(err_msg) else: 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) return (tokens[0], tokens[1], int(tokens[2])) def _prepare_ref(self, ref: AnyStr) -> bytes: # Required for command to separate refs on stdin, as bytes. if isinstance(ref, bytes): # Assume 40 bytes hexsha - bin-to-ascii for some reason returns bytes, not text. refstr: str = ref.decode("ascii") elif not isinstance(ref, str): refstr = str(ref) # Could be ref-object. else: refstr = ref if not refstr.endswith("\n"): refstr += "\n" return refstr.encode(defenc) def _get_persistent_cmd(self, attr_name: str, cmd_name: str, *args: Any, **kwargs: Any) -> "Git.AutoInterrupt": 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) cmd = cast("Git.AutoInterrupt", cmd) return cmd def __get_object_header(self, cmd: "Git.AutoInterrupt", ref: AnyStr) -> Tuple[str, str, int]: if cmd.stdin and cmd.stdout: cmd.stdin.write(self._prepare_ref(ref)) cmd.stdin.flush() return self._parse_object_header(cmd.stdout.readline()) else: raise ValueError("cmd stdin was empty") def get_object_header(self, ref: str) -> Tuple[str, str, int]: """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: str) -> Tuple[str, str, int, bytes]: """Similar to :meth:`get_object_header`, but returns object data as well. :return: (hexsha, type_string, size_as_int, data_string) :note: Not threadsafe. """ hexsha, typename, size, stream = self.stream_object_data(ref) data = stream.read(size) del stream return (hexsha, typename, size, data) def stream_object_data(self, ref: str) -> Tuple[str, str, int, "Git.CatFileContentStream"]: """Similar to :meth:`get_object_data`, but returns the data as a stream. :return: (hexsha, type_string, size_as_int, stream) :note: This method is not threadsafe. You need one independent :class:`Git` instance per thread to be safe! """ cmd = self._get_persistent_cmd("cat_file_all", "cat_file", batch=True) hexsha, typename, size = self.__get_object_header(cmd, ref) cmd_stdout = cmd.stdout if cmd.stdout is not None else io.BytesIO() return (hexsha, typename, size, self.CatFileContentStream(size, cmd_stdout)) def clear_cache(self) -> "Git": """Clear all kinds of internal caches to release resources. Currently persistent commands will be interrupted. :return: self """ for cmd in (self.cat_file_all, self.cat_file_header): if cmd: cmd.__del__() self.cat_file_all = None self.cat_file_header = None return self

Links:


URLs of crawlers that visited me.