diff --git a/Doc/library/filesys.rst b/Doc/library/filesys.rst --- a/Doc/library/filesys.rst +++ b/Doc/library/filesys.rst @@ -12,6 +12,7 @@ in this chapter is: .. toctree:: + pathlib.rst os.path.rst fileinput.rst stat.rst diff --git a/Doc/library/glob.rst b/Doc/library/glob.rst --- a/Doc/library/glob.rst +++ b/Doc/library/glob.rst @@ -25,6 +25,10 @@ For a literal match, wrap the meta-chara For example, ``'[?]'`` matches the character ``'?'``. +.. seealso:: + The :mod:`pathlib` module offers high-level path objects. + + .. function:: glob(pathname) Return a possibly-empty list of path names that match *pathname*, which must be diff --git a/Doc/library/os.path.rst b/Doc/library/os.path.rst --- a/Doc/library/os.path.rst +++ b/Doc/library/os.path.rst @@ -22,6 +22,11 @@ Functions such as :func:`expanduser` and explicitly when an application desires shell-like path expansion. (See also the :mod:`glob` module.) + +.. seealso:: + The :mod:`pathlib` module offers high-level path objects. + + .. note:: All of these functions accept either only bytes or only string objects as diff --git a/Doc/library/pathlib.rst b/Doc/library/pathlib.rst new file mode 100644 --- /dev/null +++ b/Doc/library/pathlib.rst @@ -0,0 +1,874 @@ + +:mod:`pathlib` --- Object-oriented filesystem paths +=================================================== + +.. module:: pathlib + :synopsis: Object-oriented filesystem paths + +.. index:: single: path; operations + +.. versionadded:: 3.4 + + +This module offers classes representing filesystem paths with semantics +appropriate for different operating systems. Path classes are divided +between :ref:`pure paths `, which provide purely computational +operations without I/O, and :ref:`concrete paths `, which +inherit from pure paths but also provide I/O operations. + +The main point of entry is the :class:`Path` class, which will instantiate +a :ref:`concrete path ` for the current platform. + +.. note:: + This module module has been included in the standard library on a + :term:`provisional basis `. Backwards incompatible + changes (up to and including removal of the package) may occur if deemed + necessary by the core developers. + +.. seealso:: + :pep:`428`: The pathlib module -- object-oriented filesystem paths. + +.. seealso:: + For low-level path manipulation on strings, you can also use the + :mod:`os.path` module. + + +Basic use +--------- + +Importing the main class:: + + >>> from pathlib import Path + +Listing subdirectories:: + + >>> p = Path('.') + >>> [x for x in p.iterdir() if x.is_dir()] + [PosixPath('.hg'), PosixPath('docs'), PosixPath('dist'), + PosixPath('__pycache__'), PosixPath('build')] + +Listing Python source files in this directory tree:: + + >>> list(p.glob('**/*.py')) + [PosixPath('test_pathlib.py'), PosixPath('setup.py'), + PosixPath('pathlib.py'), PosixPath('docs/conf.py'), + PosixPath('build/lib/pathlib.py')] + +Navigating inside a directory tree:: + + >>> p = Path('/etc') + >>> q = p / 'init.d' / 'reboot' + >>> q + PosixPath('/etc/init.d/reboot') + >>> q.resolve() + PosixPath('/etc/rc.d/init.d/halt') + +Querying path properties:: + + >>> q.exists() + True + >>> q.is_dir() + False + +Opening a file:: + + >>> with q.open() as f: f.readline() + ... + '#!/bin/bash\n' + + +.. _pure-paths: + +Pure paths +---------- + +Pure path objects provide path-handling operations which don't actually +access a filesystem. There are three ways to access these classes, which +we also call *flavours*: + + +.. class:: PurePosixPath + + A subclass of :class:`PurePath`, this path flavour represents non-Windows + filesystem paths:: + + >>> PurePosixPath('/etc') + PurePosixPath('/etc') + +.. class:: PureWindowsPath + + A subclass of :class:`PurePath`, this path flavour represents Windows + filesystem paths:: + + >>> PureWindowsPath('c:/Program Files/') + PureWindowsPath('c:/Program Files') + +.. class:: PurePath + + A generic class that represents the system's path flavour (instantiating + it creates either a :class:`PurePosixPath` or a :class:`PureWindowsPath`):: + + >>> PurePath('setup.py') + PurePosixPath('setup.py') + + +Regardless of the system you're running on, you can instantiate all of +these classes, since they don't provide any operation that does system calls. + + +Constructing paths +^^^^^^^^^^^^^^^^^^ + +Path constructors accept an arbitrary number of positional arguments. +When called without any argument, a path object points to the current +directory:: + + >>> PurePath() + PurePosixPath('.') + +Any argument can be a string or bytes object representing an arbitrary number +of path segments, but it can also be another path object:: + + >>> PurePath('foo', 'some/path', 'bar') + PurePosixPath('foo/some/path/bar') + >>> PurePath(Path('foo'), Path('bar')) + PurePosixPath('foo/bar') + +When several absolute paths are given, the last is taken as an anchor +(mimicking :func:`os.path.join`'s behaviour):: + + >>> PurePath('/etc', '/usr', 'lib64') + PurePosixPath('/usr/lib64') + >>> PureWindowsPath('c:/Windows', 'd:bar') + PureWindowsPath('d:bar') + +However, in a Windows path, changing the local root doesn't discard the +previous drive setting:: + + >>> PureWindowsPath('c:/Windows', '/Program Files') + PureWindowsPath('c:/Program Files') + +Spurious slashes and single dots are collapsed, but double dots (``'..'``) +are not, since this would change the meaning of a path in the face of +symbolic links:: + + >>> PurePath('foo//bar') + PurePosixPath('foo/bar') + >>> PurePath('foo/./bar') + PurePosixPath('foo/bar') + >>> PurePath('foo/../bar') + PurePosixPath('foo/../bar') + +(a naïve approach would make ``PurePosixPath('foo/../bar')`` equivalent +to ``PurePosixPath('bar')``, which is wrong if ``foo`` is a symbolic link +to another directory) + + +General properties +^^^^^^^^^^^^^^^^^^ + +Paths are immutable and hashable. Paths of a same flavour are comparable +and orderable. These properties respect the flavour's case-folding +semantics:: + + >>> PurePosixPath('foo') == PurePosixPath('FOO') + False + >>> PureWindowsPath('foo') == PureWindowsPath('FOO') + True + >>> PureWindowsPath('FOO') in { PureWindowsPath('foo') } + True + >>> PureWindowsPath('C:') < PureWindowsPath('d:') + True + +Paths of a different flavour compare unequal and cannot be ordered:: + + >>> PureWindowsPath('foo') == PurePosixPath('foo') + False + >>> PureWindowsPath('foo') < PurePosixPath('foo') + Traceback (most recent call last): + File "", line 1, in + TypeError: unorderable types: PureWindowsPath() < PurePosixPath() + + +Operators +^^^^^^^^^ + +The slash operator helps create child paths, similarly to :func:`os.path.join`:: + + >>> p = PurePath('/etc') + >>> p + PurePosixPath('/etc') + >>> p / 'init.d' / 'apache2' + PurePosixPath('/etc/init.d/apache2') + >>> q = PurePath('bin') + >>> '/usr' / q + PurePosixPath('/usr/bin') + +The string representation of a path is the raw filesystem path itself +(in native form, e.g. with backslashes under Windows), which you can +pass to any function taking a file path as a string:: + + >>> p = PurePath('/etc') + >>> str(p) + '/etc' + >>> p = PureWindowsPath('c:/Program Files') + >>> str(p) + 'c:\\Program Files' + +Similarly, calling :class:`bytes` on a path gives the raw filesystem path as a +bytes object, as encoded by :func:`os.fsencode`:: + + >>> bytes(p) + b'/etc' + +.. note:: + Calling :class:`bytes` is only recommended under Unix. Under Windows, + the unicode form is the canonical representation of filesystem paths. + + +Accessing individual parts +^^^^^^^^^^^^^^^^^^^^^^^^^^ + +To access the individual "parts" (components) of a path, use the following +property: + +.. data:: PurePath.parts + + A tuple giving access to the path's various components:: + + >>> p = PurePath('/usr/bin/python3') + >>> p.parts + ('/', 'usr', 'bin', 'python3') + + >>> p = PureWindowsPath('c:/Program Files/PSF') + >>> p.parts + ('c:\\', 'Program Files', 'PSF') + + (note how the drive and local root are regrouped in a single part) + + +Methods and properties +^^^^^^^^^^^^^^^^^^^^^^ + +Pure paths provide the following methods an properties: + +.. data:: PurePath.drive + + A string representing the drive letter or name, if any:: + + >>> PureWindowsPath('c:/Program Files/').drive + 'c:' + >>> PureWindowsPath('/Program Files/').drive + '' + >>> PurePosixPath('/etc').drive + '' + + UNC shares are also considered drives:: + + >>> PureWindowsPath('//host/share/foo.txt').drive + '\\\\host\\share' + +.. data:: PurePath.root + + A string representing the (local or global) root, if any:: + + >>> PureWindowsPath('c:/Program Files/').root + '\\' + >>> PureWindowsPath('c:Program Files/').root + '' + >>> PurePosixPath('/etc').root + '/' + + UNC shares always have a root:: + + >>> PureWindowsPath('//host/share').root + '\\' + +.. data:: PurePath.anchor + + The concatenation of the drive and root:: + + >>> PureWindowsPath('c:/Program Files/').anchor + 'c:\\' + >>> PureWindowsPath('c:Program Files/').anchor + 'c:' + >>> PurePosixPath('/etc').anchor + '/' + >>> PureWindowsPath('//host/share').anchor + '\\\\host\\share\\' + + +.. data:: PurePath.parents + + An immutable sequence providing access to the logical ancestors of + the path:: + + >>> p = PureWindowsPath('c:/foo/bar/setup.py') + >>> p.parents[0] + PureWindowsPath('c:/foo/bar') + >>> p.parents[1] + PureWindowsPath('c:/foo') + >>> p.parents[2] + PureWindowsPath('c:/') + + +.. data:: PurePath.parent + + The logical parent of the path:: + + >>> p = PurePosixPath('/a/b/c/d') + >>> p.parent + PurePosixPath('/a/b/c') + + You cannot go past an anchor, or empty path:: + + >>> p = PurePosixPath('/') + >>> p.parent + PurePosixPath('/') + >>> p = PurePosixPath('.') + >>> p.parent + PurePosixPath('.') + + .. note:: + This is a purely lexical operation, hence the following behaviour:: + + >>> p = PurePosixPath('foo/..') + >>> p.parent + PurePosixPath('foo') + + If you want to walk an arbitrary filesystem path upwards, it is + recommended to first call :meth:`Path.resolve` so as to resolve + symlinks and eliminate `".."` components. + + +.. data:: PurePath.name + + A string representing the final path component, excluding the drive and + root, if any:: + + >>> PurePosixPath('my/library/setup.py').name + 'setup.py' + + UNC drive names are not considered:: + + >>> PureWindowsPath('//some/share/setup.py').name + 'setup.py' + >>> PureWindowsPath('//some/share').name + '' + + +.. data:: PurePath.suffix + + The file extension of the final component, if any:: + + >>> PurePosixPath('my/library/setup.py').suffix + '.py' + >>> PurePosixPath('my/library.tar.gz').suffix + '.gz' + >>> PurePosixPath('my/library').suffix + '' + + +.. data:: PurePath.suffixes + + A list of the path's file extensions:: + + >>> PurePosixPath('my/library.tar.gar').suffixes + ['.tar', '.gar'] + >>> PurePosixPath('my/library.tar.gz').suffixes + ['.tar', '.gz'] + >>> PurePosixPath('my/library').suffixes + [] + + +.. data:: PurePath.stem + + The final path component, without its suffix:: + + >>> PurePosixPath('my/library.tar.gz').stem + 'library.tar' + >>> PurePosixPath('my/library.tar').stem + 'library' + >>> PurePosixPath('my/library').stem + 'library' + + +.. method:: PurePath.as_posix() + + Return a string representation of the path with forward slashes (``/``):: + + >>> p = PureWindowsPath('c:\\windows') + >>> str(p) + 'c:\\windows' + >>> p.as_posix() + 'c:/windows' + + +.. method:: PurePath.as_uri() + + Represent the path as a ``file`` URI. :exc:`ValueError` is raised if + the path isn't absolute. + + >>> p = PurePosixPath('/etc/passwd') + >>> p.as_uri() + 'file:///etc/passwd' + >>> p = PureWindowsPath('c:/Windows') + >>> p.as_uri() + 'file:///c:/Windows' + + +.. method:: PurePath.is_absolute() + + Return whether the path is absolute or not. A path is considered absolute + if it has both a root and (if the flavour allows) a drive:: + + >>> PurePosixPath('/a/b').is_absolute() + True + >>> PurePosixPath('a/b').is_absolute() + False + + >>> PureWindowsPath('c:/a/b').is_absolute() + True + >>> PureWindowsPath('/a/b').is_absolute() + False + >>> PureWindowsPath('c:').is_absolute() + False + >>> PureWindowsPath('//some/share').is_absolute() + True + + +.. method:: PurePath.is_reserved() + + With :class:`PureWindowsPath`, return True if the path is considered + reserved under Windows, False otherwise. With :class:`PurePosixPath`, + False is always returned. + + >>> PureWindowsPath('nul').is_reserved() + True + >>> PurePosixPath('nul').is_reserved() + False + + File system calls on reserved paths can fail mysteriously or have + unintended effects. + + +.. method:: PurePath.joinpath(*other) + + Calling this method is equivalent to indexing the path with each of + the *other* arguments in turn:: + + >>> PurePosixPath('/etc').joinpath('passwd') + PurePosixPath('/etc/passwd') + >>> PurePosixPath('/etc').joinpath(PurePosixPath('passwd')) + PurePosixPath('/etc/passwd') + >>> PurePosixPath('/etc').joinpath('init.d', 'apache2') + PurePosixPath('/etc/init.d/apache2') + >>> PureWindowsPath('c:').joinpath('/Program Files') + PureWindowsPath('c:/Program Files') + + +.. method:: PurePath.match(pattern) + + Match this path against the provided glob-style pattern. Return True + if matching is successful, False otherwise. + + If *pattern* is relative, the path can be either relative or absolute, + and matching is done from the right:: + + >>> PurePath('a/b.py').match('*.py') + True + >>> PurePath('/a/b/c.py').match('b/*.py') + True + >>> PurePath('/a/b/c.py').match('a/*.py') + False + + If *pattern* is absolute, the path must be absolute, and the whole path + must match:: + + >>> PurePath('/a.py').match('/*.py') + True + >>> PurePath('a/b.py').match('/*.py') + False + + As with other methods, case-sensitivity is observed:: + + >>> PureWindowsPath('b.py').match('*.PY') + True + + +.. method:: PurePath.relative_to(*other) + + Compute a version of this path relative to the path represented by + *other*. If it's impossible, ValueError is raised:: + + >>> p = PurePosixPath('/etc/passwd') + >>> p.relative_to('/') + PurePosixPath('etc/passwd') + >>> p.relative_to('/etc') + PurePosixPath('passwd') + >>> p.relative_to('/usr') + Traceback (most recent call last): + File "", line 1, in + File "pathlib.py", line 694, in relative_to + .format(str(self), str(formatted))) + ValueError: '/etc/passwd' does not start with '/usr' + + +.. _concrete-paths: + + +Concrete paths +-------------- + +Concrete paths are subclasses of the pure path classes. In addition to +operations provided by the latter, they also provide methods to do system +calls on path objects. There are three ways to instantiate concrete paths: + + +.. class:: PosixPath + + A subclass of :class:`Path` and :class:`PurePosixPath`, this class + represents concrete non-Windows filesystem paths:: + + >>> PosixPath('/etc') + PosixPath('/etc') + +.. class:: WindowsPath + + A subclass of :class:`Path` and :class:`PureWindowsPath`, this class + represents concrete Windows filesystem paths:: + + >>> WindowsPath('c:/Program Files/') + WindowsPath('c:/Program Files') + +.. class:: Path + + A subclass of :class:`PurePath`, this class represents concrete paths of + the system's path flavour (instantiating it creates either a + :class:`PosixPath` or a :class:`WindowsPath`):: + + >>> Path('setup.py') + PosixPath('setup.py') + + +You can only instantiate the class flavour that corresponds to your system +(allowing system calls on non-compatible path flavours could lead to +bugs or failures in your application):: + + >>> import os + >>> os.name + 'posix' + >>> Path('setup.py') + PosixPath('setup.py') + >>> PosixPath('setup.py') + PosixPath('setup.py') + >>> WindowsPath('setup.py') + Traceback (most recent call last): + File "", line 1, in + File "pathlib.py", line 798, in __new__ + % (cls.__name__,)) + NotImplementedError: cannot instantiate 'WindowsPath' on your system + + +Methods +^^^^^^^ + +Concrete paths provide the following methods in addition to pure paths +methods. Many of these methods can raise an :exc:`OSError` if a system +call fails (for example because the path doesn't exist): + +.. classmethod:: Path.cwd() + + Return a new path object representing the current directory (as returned + by :func:`os.getcwd`):: + + >>> Path.cwd() + PosixPath('/home/antoine/pathlib') + + +.. method:: Path.stat() + + Return information about this path (similarly to :func:`os.stat`). + The result is looked up at each call to this method. + + >>> p = Path('setup.py') + >>> p.stat().st_size + 956 + >>> p.stat().st_mtime + 1327883547.852554 + + +.. method:: Path.chmod(mode) + + Change the file mode and permissions, like :func:`os.chmod`:: + + >>> p = Path('setup.py') + >>> p.stat().st_mode + 33277 + >>> p.chmod(0o444) + >>> p.stat().st_mode + 33060 + + +.. method:: Path.exists() + + Whether the path points to an existing file or directory:: + + >>> Path('.').exists() + True + >>> Path('setup.py').exists() + True + >>> Path('/etc').exists() + True + >>> Path('nonexistentfile').exists() + False + + .. note:: + If the path points to a symlink, :meth:`exists` returns whether the + symlink *points to* an existing file or directory. + + +.. method:: Path.glob(pattern) + + Glob the given *pattern* in the directory represented by this path, + yielding all matching files (of any kind):: + + >>> sorted(Path('.').glob('*.py')) + [PosixPath('pathlib.py'), PosixPath('setup.py'), PosixPath('test_pathlib.py')] + >>> sorted(Path('.').glob('*/*.py')) + [PosixPath('docs/conf.py')] + + The "``**``" pattern means "this directory and all subdirectories, + recursively". In other words, it enables recursive globbing:: + + >>> sorted(Path('.').glob('**/*.py')) + [PosixPath('build/lib/pathlib.py'), + PosixPath('docs/conf.py'), + PosixPath('pathlib.py'), + PosixPath('setup.py'), + PosixPath('test_pathlib.py')] + + .. note:: + Using the "``**``" pattern in large directory trees may consume + an inordinate amount of time. + + +.. method:: Path.group() + + Return the name of the group owning the file. :exc:`KeyError` is thrown + if the file's gid isn't found in the system database. + + +.. method:: Path.is_dir() + + Return True if the path points to a directory (or a symbolic link + pointing to a directory), False if it points to another kind of file. + + False is also returned if the path doesn't exist or is a broken symlink; + other errors (such as permission errors) are propagated. + + +.. method:: Path.is_file() + + Return True if the path points to a regular file (or a symbolic link + pointing to a regular file), False if it points to another kind of file. + + False is also returned if the path doesn't exist or is a broken symlink; + other errors (such as permission errors) are propagated. + + +.. method:: Path.is_symlink() + + Return True if the path points to a symbolic link, False otherwise. + + False is also returned if the path doesn't exist; other errors (such + as permission errors) are propagated. + + +.. method:: Path.is_socket() + + Return True if the path points to a Unix socket (or a symbolic link + pointing to a Unix socket), False if it points to another kind of file. + + False is also returned if the path doesn't exist or is a broken symlink; + other errors (such as permission errors) are propagated. + + +.. method:: Path.is_fifo() + + Return True if the path points to a FIFO (or a symbolic link + pointing to a FIFO), False if it points to another kind of file. + + False is also returned if the path doesn't exist or is a broken symlink; + other errors (such as permission errors) are propagated. + + +.. method:: Path.is_block_device() + + Return True if the path points to a block device (or a symbolic link + pointing to a block device), False if it points to another kind of file. + + False is also returned if the path doesn't exist or is a broken symlink; + other errors (such as permission errors) are propagated. + + +.. method:: Path.is_char_device() + + Return True if the path points to a character device (or a symbolic link + pointing to a character device), False if it points to another kind of file. + + False is also returned if the path doesn't exist or is a broken symlink; + other errors (such as permission errors) are propagated. + + +.. method:: Path.iterdir() + + When the path points to a directory, yield path objects of the directory + contents:: + + >>> p = Path('docs') + >>> for child in p.iterdir(): child + ... + PosixPath('docs/conf.py') + PosixPath('docs/_templates') + PosixPath('docs/make.bat') + PosixPath('docs/index.rst') + PosixPath('docs/_build') + PosixPath('docs/_static') + PosixPath('docs/Makefile') + +.. method:: Path.lchmod(mode) + + Like :meth:`Path.chmod` but, if the path points to a symbolic link, the + symbolic link's mode is changed rather than its target's. + + +.. method:: Path.lstat() + + Like :meth:`Path.stat` but, if the path points to a symbolic link, return + the symbolic link's information rather than its target's. + + +.. method:: Path.mkdir(mode=0o777, parents=False) + + Create a new directory at this given path. If *mode* is given, it is + combined with the process' ``umask`` value to determine the file mode + and access flags. If the path already exists, :exc:`OSError` is raised. + + If *parents* is True, any missing parents of this path are created + as needed. If *parents* is False (the default), a missing parent raises + :exc:`OSError`. + + +.. method:: Path.open(mode='r', buffering=-1, encoding=None, errors=None, newline=None) + + Open the file pointed to by the path, like the built-in :func:`open` + function does:: + + >>> p = Path('setup.py') + >>> with p.open() as f: + ... f.readline() + ... + '#!/usr/bin/env python3\n' + + +.. method:: Path.owner() + + Return the name of the user owning the file. :exc:`KeyError` is thrown + if the file's uid isn't found in the system database. + + +.. method:: Path.rename(target) + + Rename this file or directory to the given *target*. *target* can be + either a string or another path object:: + + >>> p = Path('foo') + >>> p.open('w').write('some text') + 9 + >>> target = Path('bar') + >>> p.rename(target) + >>> target.open().read() + 'some text' + + +.. method:: Path.replace(target) + + Rename this file or directory to the given *target*. If *target* points + to an existing file or directory, it will be unconditionally replaced. + + +.. method:: Path.resolve() + + Make the path absolute, resolving any symlinks. A new path object is + returned:: + + >>> p = Path() + >>> p + PosixPath('.') + >>> p.resolve() + PosixPath('/home/antoine/pathlib') + + `".."` components are also eliminated (this is the only method to do so):: + + >>> p = Path('docs/../setup.py') + >>> p.resolve() + PosixPath('/home/antoine/pathlib/setup.py') + + If the path doesn't exist, :exc:`FileNotFoundError` is raised. If an + infinite loop is encountered along the resolution path, + :exc:`RuntimeError` is raised. + + +.. method:: Path.rglob(pattern) + + This is like calling :meth:`glob` with "``**``" added in front of the + given *pattern*: + + >>> sorted(Path().rglob("*.py")) + [PosixPath('build/lib/pathlib.py'), + PosixPath('docs/conf.py'), + PosixPath('pathlib.py'), + PosixPath('setup.py'), + PosixPath('test_pathlib.py')] + + +.. method:: Path.rmdir() + + Remove this directory. The directory must be empty. + + +.. method:: Path.symlink_to(target, target_is_directory=False) + + Make this path a symbolic link to *target*. Under Windows, + *target_is_directory* must be True (default False) if the link's target + is a directory. Under POSIX, *target_is_directory*'s value is ignored. + + >>> p = Path('mylink') + >>> p.symlink_to('setup.py') + >>> p.resolve() + PosixPath('/home/antoine/pathlib/setup.py') + >>> p.stat().st_size + 956 + >>> p.lstat().st_size + 8 + + .. note:: + The order of arguments (link, target) is the reverse + of :func:`os.symlink`'s. + + +.. method:: Path.touch(mode=0o777, exist_ok=True) + + Create a file at this given path. If *mode* is given, it is combined + with the process' ``umask`` value to determine the file mode and access + flags. If the file already exists, the function succeeds if *exist_ok* + is true (and its modification time is updated to the current time), + otherwise :exc:`OSError` is raised. + + +.. method:: Path.unlink() + + Remove this file or symbolic link. If the path points to a directory, + use :func:`Path.rmdir` instead. + diff --git a/Doc/whatsnew/3.4.rst b/Doc/whatsnew/3.4.rst --- a/Doc/whatsnew/3.4.rst +++ b/Doc/whatsnew/3.4.rst @@ -93,6 +93,7 @@ New library modules: * :mod:`asyncio`: New provisonal API for asynchronous IO (:pep:`3156`). * :mod:`enum`: Support for enumeration types (:pep:`435`). * :mod:`ensurepip`: Bootstrapping the pip installer (:pep:`453`). +* :mod:`pathlib`: Object-oriented filesystem paths (:pep:`428`). * :mod:`selectors`: High-level and efficient I/O multiplexing, built upon the :mod:`select` module primitives. * :mod:`statistics`: A basic numerically stable statistics library (:pep:`450`). @@ -298,6 +299,21 @@ compatible enumeration values. implemented by Ethan Furman. +pathlib +------- + +The new :mod:`pathlib` module offers classes representing filesystem paths +with semantics appropriate for different operating systems. Path classes are +divided between *pure paths*, which provide purely computational operations +without I/O, and *concrete paths*, which inherit from pure paths but also +provide I/O operations. + +.. seealso:: + + :pep:`428` - The pathlib module -- object-oriented filesystem paths + PEP written and implemented by Antoine Pitrou. + + selectors --------- diff --git a/Lib/pathlib.py b/Lib/pathlib.py new file mode 100644 --- /dev/null +++ b/Lib/pathlib.py @@ -0,0 +1,1287 @@ +import fnmatch +import functools +import io +import ntpath +import os +import posixpath +import re +import sys +import time +import weakref +try: + import threading +except ImportError: + import dummy_threading as threading + +from collections import Sequence, defaultdict +from contextlib import contextmanager +from errno import EINVAL, ENOENT +from itertools import chain, count +from operator import attrgetter +from stat import S_ISDIR, S_ISLNK, S_ISREG, S_ISSOCK, S_ISBLK, S_ISCHR, S_ISFIFO +from urllib.parse import quote as urlquote, quote_from_bytes as urlquote_from_bytes + + +supports_symlinks = True +try: + import nt +except ImportError: + nt = None +else: + if sys.getwindowsversion()[:2] >= (6, 0): + from nt import _getfinalpathname + else: + supports_symlinks = False + _getfinalpathname = None + + +__all__ = [ + "PurePath", "PurePosixPath", "PureWindowsPath", + "Path", "PosixPath", "WindowsPath", + ] + +# +# Internals +# + +def _is_wildcard_pattern(pat): + # Whether this pattern needs actual matching using fnmatch, or can + # be looked up directly as a file. + return "*" in pat or "?" in pat or "[" in pat + + +class _Flavour(object): + """A flavour implements a particular (platform-specific) set of path + semantics.""" + + def __init__(self): + self.join = self.sep.join + + def parse_parts(self, parts): + parsed = [] + sep = self.sep + altsep = self.altsep + drv = root = '' + it = reversed(parts) + for part in it: + if not part: + continue + if altsep: + part = part.replace(altsep, sep) + drv, root, rel = self.splitroot(part) + if sep in rel: + for x in reversed(rel.split(sep)): + if x and x != '.': + parsed.append(sys.intern(x)) + else: + if rel and rel != '.': + parsed.append(sys.intern(rel)) + if drv or root: + if not drv: + # If no drive is present, try to find one in the previous + # parts. This makes the result of parsing e.g. + # ("C:", "/", "a") reasonably intuitive. + for part in it: + drv = self.splitroot(part)[0] + if drv: + break + break + if drv or root: + parsed.append(drv + root) + parsed.reverse() + return drv, root, parsed + + def join_parsed_parts(self, drv, root, parts, drv2, root2, parts2): + """ + Join the two paths represented by the respective + (drive, root, parts) tuples. Return a new (drive, root, parts) tuple. + """ + if root2: + parts = parts2 + root = root2 + else: + parts = parts + parts2 + # XXX raise error if drv and drv2 are different? + return drv2 or drv, root, parts + + +class _WindowsFlavour(_Flavour): + # Reference for Windows paths can be found at + # http://msdn.microsoft.com/en-us/library/aa365247%28v=vs.85%29.aspx + + sep = '\\' + altsep = '/' + has_drv = True + pathmod = ntpath + + is_supported = (nt is not None) + + drive_letters = ( + set(chr(x) for x in range(ord('a'), ord('z') + 1)) | + set(chr(x) for x in range(ord('A'), ord('Z') + 1)) + ) + ext_namespace_prefix = '\\\\?\\' + + reserved_names = ( + {'CON', 'PRN', 'AUX', 'NUL'} | + {'COM%d' % i for i in range(1, 10)} | + {'LPT%d' % i for i in range(1, 10)} + ) + + # Interesting findings about extended paths: + # - '\\?\c:\a', '//?/c:\a' and '//?/c:/a' are all supported + # but '\\?\c:/a' is not + # - extended paths are always absolute; "relative" extended paths will + # fail. + + def splitroot(self, part, sep=sep): + first = part[0:1] + second = part[1:2] + if (second == sep and first == sep): + # XXX extended paths should also disable the collapsing of "." + # components (according to MSDN docs). + prefix, part = self._split_extended_path(part) + first = part[0:1] + second = part[1:2] + else: + prefix = '' + third = part[2:3] + if (second == sep and first == sep and third != sep): + # is a UNC path: + # vvvvvvvvvvvvvvvvvvvvv root + # \\machine\mountpoint\directory\etc\... + # directory ^^^^^^^^^^^^^^ + index = part.find(sep, 2) + if index != -1: + index2 = part.find(sep, index + 1) + # a UNC path can't have two slashes in a row + # (after the initial two) + if index2 != index + 1: + if index2 == -1: + index2 = len(part) + if prefix: + return prefix + part[1:index2], sep, part[index2+1:] + else: + return part[:index2], sep, part[index2+1:] + drv = root = '' + if second == ':' and first in self.drive_letters: + drv = part[:2] + part = part[2:] + first = third + if first == sep: + root = first + part = part.lstrip(sep) + return prefix + drv, root, part + + def casefold(self, s): + return s.lower() + + def casefold_parts(self, parts): + return [p.lower() for p in parts] + + def resolve(self, path): + s = str(path) + if not s: + return os.getcwd() + if _getfinalpathname is not None: + return self._ext_to_normal(_getfinalpathname(s)) + # Means fallback on absolute + return None + + def _split_extended_path(self, s, ext_prefix=ext_namespace_prefix): + prefix = '' + if s.startswith(ext_prefix): + prefix = s[:4] + s = s[4:] + if s.startswith('UNC\\'): + prefix += s[:3] + s = '\\' + s[3:] + return prefix, s + + def _ext_to_normal(self, s): + # Turn back an extended path into a normal DOS-like path + return self._split_extended_path(s)[1] + + def is_reserved(self, parts): + # NOTE: the rules for reserved names seem somewhat complicated + # (e.g. r"..\NUL" is reserved but not r"foo\NUL"). + # We err on the side of caution and return True for paths which are + # not considered reserved by Windows. + if not parts: + return False + if parts[0].startswith('\\\\'): + # UNC paths are never reserved + return False + return parts[-1].partition('.')[0].upper() in self.reserved_names + + def make_uri(self, path): + # Under Windows, file URIs use the UTF-8 encoding. + drive = path.drive + if len(drive) == 2 and drive[1] == ':': + # It's a path on a local drive => 'file:///c:/a/b' + rest = path.as_posix()[2:].lstrip('/') + return 'file:///%s/%s' % ( + drive, urlquote_from_bytes(rest.encode('utf-8'))) + else: + # It's a path on a network drive => 'file://host/share/a/b' + return 'file:' + urlquote_from_bytes(path.as_posix().encode('utf-8')) + + +class _PosixFlavour(_Flavour): + sep = '/' + altsep = '' + has_drv = False + pathmod = posixpath + + is_supported = (os.name != 'nt') + + def splitroot(self, part, sep=sep): + if part and part[0] == sep: + stripped_part = part.lstrip(sep) + # According to POSIX path resolution: + # http://pubs.opengroup.org/onlinepubs/009695399/basedefs/xbd_chap04.html#tag_04_11 + # "A pathname that begins with two successive slashes may be + # interpreted in an implementation-defined manner, although more + # than two leading slashes shall be treated as a single slash". + if len(part) - len(stripped_part) == 2: + return '', sep * 2, stripped_part + else: + return '', sep, stripped_part + else: + return '', '', part + + def casefold(self, s): + return s + + def casefold_parts(self, parts): + return parts + + def resolve(self, path): + sep = self.sep + def split(p): + return [x for x in p.split(sep) if x] + def absparts(p): + # Our own abspath(), since the posixpath one makes + # the mistake of "normalizing" the path without resolving the + # symlinks first. + if not p.startswith(sep): + return split(os.getcwd()) + split(p) + else: + return split(p) + parts = absparts(str(path))[::-1] + accessor = path._accessor + resolved = cur = "" + symlinks = {} + while parts: + part = parts.pop() + cur = resolved + sep + part + if cur in symlinks and symlinks[cur] <= len(parts): + # We've already seen the symlink and there's not less + # work to do than the last time. + raise RuntimeError("Symlink loop from %r" % cur) + try: + target = accessor.readlink(cur) + except OSError as e: + if e.errno != EINVAL: + raise + # Not a symlink + resolved = cur + else: + # Take note of remaining work from this symlink + symlinks[cur] = len(parts) + if target.startswith(sep): + # Symlink points to absolute path + resolved = "" + parts.extend(split(target)[::-1]) + return resolved or sep + + def is_reserved(self, parts): + return False + + def make_uri(self, path): + # We represent the path using the local filesystem encoding, + # for portability to other applications. + bpath = bytes(path) + return 'file://' + urlquote_from_bytes(bpath) + + +_windows_flavour = _WindowsFlavour() +_posix_flavour = _PosixFlavour() + + +class _Accessor: + """An accessor implements a particular (system-specific or not) way of + accessing paths on the filesystem.""" + + +class _NormalAccessor(_Accessor): + + def _wrap_strfunc(strfunc): + @functools.wraps(strfunc) + def wrapped(pathobj, *args): + return strfunc(str(pathobj), *args) + return staticmethod(wrapped) + + def _wrap_binary_strfunc(strfunc): + @functools.wraps(strfunc) + def wrapped(pathobjA, pathobjB, *args): + return strfunc(str(pathobjA), str(pathobjB), *args) + return staticmethod(wrapped) + + stat = _wrap_strfunc(os.stat) + + lstat = _wrap_strfunc(os.lstat) + + open = _wrap_strfunc(os.open) + + listdir = _wrap_strfunc(os.listdir) + + chmod = _wrap_strfunc(os.chmod) + + if hasattr(os, "lchmod"): + lchmod = _wrap_strfunc(os.lchmod) + else: + def lchmod(self, pathobj, mode): + raise NotImplementedError("lchmod() not available on this system") + + mkdir = _wrap_strfunc(os.mkdir) + + unlink = _wrap_strfunc(os.unlink) + + rmdir = _wrap_strfunc(os.rmdir) + + rename = _wrap_binary_strfunc(os.rename) + + replace = _wrap_binary_strfunc(os.replace) + + if nt: + if supports_symlinks: + symlink = _wrap_binary_strfunc(os.symlink) + else: + def symlink(a, b, target_is_directory): + raise NotImplementedError("symlink() not available on this system") + else: + # Under POSIX, os.symlink() takes two args + @staticmethod + def symlink(a, b, target_is_directory): + return os.symlink(str(a), str(b)) + + utime = _wrap_strfunc(os.utime) + + # Helper for resolve() + def readlink(self, path): + return os.readlink(path) + + +_normal_accessor = _NormalAccessor() + + +# +# Globbing helpers +# + +@contextmanager +def _cached(func): + try: + func.__cached__ + yield func + except AttributeError: + cache = {} + def wrapper(*args): + try: + return cache[args] + except KeyError: + value = cache[args] = func(*args) + return value + wrapper.__cached__ = True + try: + yield wrapper + finally: + cache.clear() + +def _make_selector(pattern_parts): + pat = pattern_parts[0] + child_parts = pattern_parts[1:] + if pat == '**': + cls = _RecursiveWildcardSelector + elif '**' in pat: + raise ValueError("Invalid pattern: '**' can only be an entire path component") + elif _is_wildcard_pattern(pat): + cls = _WildcardSelector + else: + cls = _PreciseSelector + return cls(pat, child_parts) + +if hasattr(functools, "lru_cache"): + _make_selector = functools.lru_cache()(_make_selector) + + +class _Selector: + """A selector matches a specific glob pattern part against the children + of a given path.""" + + def __init__(self, child_parts): + self.child_parts = child_parts + if child_parts: + self.successor = _make_selector(child_parts) + else: + self.successor = _TerminatingSelector() + + def select_from(self, parent_path): + """Iterate over all child paths of `parent_path` matched by this + selector. This can contain parent_path itself.""" + path_cls = type(parent_path) + is_dir = path_cls.is_dir + exists = path_cls.exists + listdir = parent_path._accessor.listdir + return self._select_from(parent_path, is_dir, exists, listdir) + + +class _TerminatingSelector: + + def _select_from(self, parent_path, is_dir, exists, listdir): + yield parent_path + + +class _PreciseSelector(_Selector): + + def __init__(self, name, child_parts): + self.name = name + _Selector.__init__(self, child_parts) + + def _select_from(self, parent_path, is_dir, exists, listdir): + if not is_dir(parent_path): + return + path = parent_path._make_child_relpath(self.name) + if exists(path): + for p in self.successor._select_from(path, is_dir, exists, listdir): + yield p + + +class _WildcardSelector(_Selector): + + def __init__(self, pat, child_parts): + self.pat = re.compile(fnmatch.translate(pat)) + _Selector.__init__(self, child_parts) + + def _select_from(self, parent_path, is_dir, exists, listdir): + if not is_dir(parent_path): + return + cf = parent_path._flavour.casefold + for name in listdir(parent_path): + casefolded = cf(name) + if self.pat.match(casefolded): + path = parent_path._make_child_relpath(name) + for p in self.successor._select_from(path, is_dir, exists, listdir): + yield p + + +class _RecursiveWildcardSelector(_Selector): + + def __init__(self, pat, child_parts): + _Selector.__init__(self, child_parts) + + def _iterate_directories(self, parent_path, is_dir, listdir): + yield parent_path + for name in listdir(parent_path): + path = parent_path._make_child_relpath(name) + if is_dir(path): + for p in self._iterate_directories(path, is_dir, listdir): + yield p + + def _select_from(self, parent_path, is_dir, exists, listdir): + if not is_dir(parent_path): + return + with _cached(listdir) as listdir: + yielded = set() + try: + successor_select = self.successor._select_from + for starting_point in self._iterate_directories(parent_path, is_dir, listdir): + for p in successor_select(starting_point, is_dir, exists, listdir): + if p not in yielded: + yield p + yielded.add(p) + finally: + yielded.clear() + + +# +# Public API +# + +class _PathParents(Sequence): + """This object provides sequence-like access to the logical ancestors + of a path. Don't try to construct it yourself.""" + __slots__ = ('_pathcls', '_drv', '_root', '_parts') + + def __init__(self, path): + # We don't store the instance to avoid reference cycles + self._pathcls = type(path) + self._drv = path._drv + self._root = path._root + self._parts = path._parts + + def __len__(self): + if self._drv or self._root: + return len(self._parts) - 1 + else: + return len(self._parts) + + def __getitem__(self, idx): + if idx < 0 or idx >= len(self): + raise IndexError(idx) + return self._pathcls._from_parsed_parts(self._drv, self._root, + self._parts[:-idx - 1]) + + def __repr__(self): + return "<{}.parents>".format(self._pathcls.__name__) + + +class PurePath(object): + """PurePath represents a filesystem path and offers operations which + don't imply any actual filesystem I/O. Depending on your system, + instantiating a PurePath will return either a PurePosixPath or a + PureWindowsPath object. You can also instantiate either of these classes + directly, regardless of your system. + """ + __slots__ = ( + '_drv', '_root', '_parts', + '_str', '_hash', '_pparts', '_cached_cparts', + ) + + def __new__(cls, *args): + """Construct a PurePath from one or several strings and or existing + PurePath objects. The strings and path objects are combined so as + to yield a canonicalized path, which is incorporated into the + new PurePath object. + """ + if cls is PurePath: + cls = PureWindowsPath if os.name == 'nt' else PurePosixPath + return cls._from_parts(args) + + def __reduce__(self): + # Using the parts tuple helps share interned path parts + # when pickling related paths. + return (self.__class__, tuple(self._parts)) + + @classmethod + def _parse_args(cls, args): + # This is useful when you don't want to create an instance, just + # canonicalize some constructor arguments. + parts = [] + for a in args: + if isinstance(a, PurePath): + parts += a._parts + elif isinstance(a, str): + # Assuming a str + parts.append(a) + else: + raise TypeError( + "argument should be a path or str object, not %r" + % type(a)) + return cls._flavour.parse_parts(parts) + + @classmethod + def _from_parts(cls, args, init=True): + # We need to call _parse_args on the instance, so as to get the + # right flavour. + self = object.__new__(cls) + drv, root, parts = self._parse_args(args) + self._drv = drv + self._root = root + self._parts = parts + if init: + self._init() + return self + + @classmethod + def _from_parsed_parts(cls, drv, root, parts, init=True): + self = object.__new__(cls) + self._drv = drv + self._root = root + self._parts = parts + if init: + self._init() + return self + + @classmethod + def _format_parsed_parts(cls, drv, root, parts): + if drv or root: + return drv + root + cls._flavour.join(parts[1:]) + else: + return cls._flavour.join(parts) + + def _init(self): + # Overriden in concrete Path + pass + + def _make_child(self, args): + drv, root, parts = self._parse_args(args) + drv, root, parts = self._flavour.join_parsed_parts( + self._drv, self._root, self._parts, drv, root, parts) + return self._from_parsed_parts(drv, root, parts) + + def __str__(self): + """Return the string representation of the path, suitable for + passing to system calls.""" + try: + return self._str + except AttributeError: + self._str = self._format_parsed_parts(self._drv, self._root, + self._parts) or '.' + return self._str + + def as_posix(self): + """Return the string representation of the path with forward (/) + slashes.""" + f = self._flavour + return str(self).replace(f.sep, '/') + + def __bytes__(self): + """Return the bytes representation of the path. This is only + recommended to use under Unix.""" + return os.fsencode(str(self)) + + def __repr__(self): + return "{}({!r})".format(self.__class__.__name__, self.as_posix()) + + def as_uri(self): + """Return the path as a 'file' URI.""" + if not self.is_absolute(): + raise ValueError("relative path can't be expressed as a file URI") + return self._flavour.make_uri(self) + + @property + def _cparts(self): + # Cached casefolded parts, for hashing and comparison + try: + return self._cached_cparts + except AttributeError: + self._cached_cparts = self._flavour.casefold_parts(self._parts) + return self._cached_cparts + + def __eq__(self, other): + if not isinstance(other, PurePath): + return NotImplemented + return self._cparts == other._cparts and self._flavour is other._flavour + + def __ne__(self, other): + return not self == other + + def __hash__(self): + try: + return self._hash + except AttributeError: + self._hash = hash(tuple(self._cparts)) + return self._hash + + def __lt__(self, other): + if not isinstance(other, PurePath) or self._flavour is not other._flavour: + return NotImplemented + return self._cparts < other._cparts + + def __le__(self, other): + if not isinstance(other, PurePath) or self._flavour is not other._flavour: + return NotImplemented + return self._cparts <= other._cparts + + def __gt__(self, other): + if not isinstance(other, PurePath) or self._flavour is not other._flavour: + return NotImplemented + return self._cparts > other._cparts + + def __ge__(self, other): + if not isinstance(other, PurePath) or self._flavour is not other._flavour: + return NotImplemented + return self._cparts >= other._cparts + + drive = property(attrgetter('_drv'), + doc="""The drive prefix (letter or UNC path), if any.""") + + root = property(attrgetter('_root'), + doc="""The root of the path, if any.""") + + @property + def anchor(self): + """The concatenation of the drive and root, or ''.""" + anchor = self._drv + self._root + return anchor + + @property + def name(self): + """The final path component, if any.""" + parts = self._parts + if len(parts) == (1 if (self._drv or self._root) else 0): + return '' + return parts[-1] + + @property + def suffix(self): + """The final component's last suffix, if any.""" + name = self.name + i = name.rfind('.') + if 0 < i < len(name) - 1: + return name[i:] + else: + return '' + + @property + def suffixes(self): + """A list of the final component's suffixes, if any.""" + name = self.name + if name.endswith('.'): + return [] + name = name.lstrip('.') + return ['.' + suffix for suffix in name.split('.')[1:]] + + @property + def stem(self): + """The final path component, minus its last suffix.""" + name = self.name + i = name.rfind('.') + if 0 < i < len(name) - 1: + return name[:i] + else: + return name + + def with_name(self, name): + """Return a new path with the file name changed.""" + if not self.name: + raise ValueError("%r has an empty name" % (self,)) + return self._from_parsed_parts(self._drv, self._root, + self._parts[:-1] + [name]) + + def with_suffix(self, suffix): + """Return a new path with the file suffix changed (or added, if none).""" + # XXX if suffix is None, should the current suffix be removed? + name = self.name + if not name: + raise ValueError("%r has an empty name" % (self,)) + old_suffix = self.suffix + if not old_suffix: + name = name + suffix + else: + name = name[:-len(old_suffix)] + suffix + return self._from_parsed_parts(self._drv, self._root, + self._parts[:-1] + [name]) + + def relative_to(self, *other): + """Return the relative path to another path identified by the passed + arguments. If the operation is not possible (because this is not + a subpath of the other path), raise ValueError. + """ + # For the purpose of this method, drive and root are considered + # separate parts, i.e.: + # Path('c:/').relative_to('c:') gives Path('/') + # Path('c:/').relative_to('/') raise ValueError + if not other: + raise TypeError("need at least one argument") + parts = self._parts + drv = self._drv + root = self._root + if drv or root: + if root: + abs_parts = [drv, root] + parts[1:] + else: + abs_parts = [drv] + parts[1:] + else: + abs_parts = parts + to_drv, to_root, to_parts = self._parse_args(other) + if to_drv or to_root: + if to_root: + to_abs_parts = [to_drv, to_root] + to_parts[1:] + else: + to_abs_parts = [to_drv] + to_parts[1:] + else: + to_abs_parts = to_parts + n = len(to_abs_parts) + if n == 0 and (drv or root) or abs_parts[:n] != to_abs_parts: + formatted = self._format_parsed_parts(to_drv, to_root, to_parts) + raise ValueError("{!r} does not start with {!r}" + .format(str(self), str(formatted))) + return self._from_parsed_parts('', '', abs_parts[n:]) + + @property + def parts(self): + """An object providing sequence-like access to the + components in the filesystem path.""" + # We cache the tuple to avoid building a new one each time .parts + # is accessed. XXX is this necessary? + try: + return self._pparts + except AttributeError: + self._pparts = tuple(self._parts) + return self._pparts + + def joinpath(self, *args): + """Combine this path with one or several arguments, and return a + new path representing either a subpath (if all arguments are relative + paths) or a totally different path (if one of the arguments is + anchored). + """ + return self._make_child(args) + + def __truediv__(self, key): + return self._make_child((key,)) + + def __rtruediv__(self, key): + return self._from_parts([key] + self._parts) + + @property + def parent(self): + """The logical parent of the path.""" + drv = self._drv + root = self._root + parts = self._parts + if len(parts) == 1 and (drv or root): + return self + return self._from_parsed_parts(drv, root, parts[:-1]) + + @property + def parents(self): + """A sequence of this path's logical parents.""" + return _PathParents(self) + + def is_absolute(self): + """True if the path is absolute (has both a root and, if applicable, + a drive).""" + if not self._root: + return False + return not self._flavour.has_drv or bool(self._drv) + + def is_reserved(self): + """Return True if the path contains one of the special names reserved + by the system, if any.""" + return self._flavour.is_reserved(self._parts) + + def match(self, path_pattern): + """ + Return True if this path matches the given pattern. + """ + cf = self._flavour.casefold + path_pattern = cf(path_pattern) + drv, root, pat_parts = self._flavour.parse_parts((path_pattern,)) + if not pat_parts: + raise ValueError("empty pattern") + if drv and drv != cf(self._drv): + return False + if root and root != cf(self._root): + return False + parts = self._cparts + if drv or root: + if len(pat_parts) != len(parts): + return False + pat_parts = pat_parts[1:] + elif len(pat_parts) > len(parts): + return False + for part, pat in zip(reversed(parts), reversed(pat_parts)): + if not fnmatch.fnmatchcase(part, pat): + return False + return True + + +class PurePosixPath(PurePath): + _flavour = _posix_flavour + __slots__ = () + + +class PureWindowsPath(PurePath): + _flavour = _windows_flavour + __slots__ = () + + +# Filesystem-accessing classes + + +class Path(PurePath): + __slots__ = ( + '_accessor', + '_closed', + ) + + def __new__(cls, *args, **kwargs): + if cls is Path: + cls = WindowsPath if os.name == 'nt' else PosixPath + self = cls._from_parts(args, init=False) + if not self._flavour.is_supported: + raise NotImplementedError("cannot instantiate %r on your system" + % (cls.__name__,)) + self._init() + return self + + def _init(self, + # Private non-constructor arguments + template=None, + ): + self._closed = False + if template is not None: + self._accessor = template._accessor + else: + self._accessor = _normal_accessor + + def _make_child_relpath(self, part): + # This is an optimization used for dir walking. `part` must be + # a single part relative to this path. + parts = self._parts + [part] + return self._from_parsed_parts(self._drv, self._root, parts) + + def __enter__(self): + if self._closed: + self._raise_closed() + return self + + def __exit__(self, t, v, tb): + self._closed = True + + def _raise_closed(self): + raise ValueError("I/O operation on closed path") + + def _opener(self, name, flags, mode=0o666): + # A stub for the opener argument to built-in open() + return self._accessor.open(self, flags, mode) + + # Public API + + @classmethod + def cwd(cls): + """Return a new path pointing to the current working directory + (as returned by os.getcwd()). + """ + return cls(os.getcwd()) + + def iterdir(self): + """Iterate over the files in this directory. Does not yield any + result for the special paths '.' and '..'. + """ + if self._closed: + self._raise_closed() + for name in self._accessor.listdir(self): + if name in {'.', '..'}: + # Yielding a path object for these makes little sense + continue + yield self._make_child_relpath(name) + if self._closed: + self._raise_closed() + + def glob(self, pattern): + """Iterate over this subtree and yield all existing files (of any + kind, including directories) matching the given pattern. + """ + pattern = self._flavour.casefold(pattern) + drv, root, pattern_parts = self._flavour.parse_parts((pattern,)) + if drv or root: + raise NotImplementedError("Non-relative patterns are unsupported") + selector = _make_selector(tuple(pattern_parts)) + for p in selector.select_from(self): + yield p + + def rglob(self, pattern): + """Recursively yield all existing files (of any kind, including + directories) matching the given pattern, anywhere in this subtree. + """ + pattern = self._flavour.casefold(pattern) + drv, root, pattern_parts = self._flavour.parse_parts((pattern,)) + if drv or root: + raise NotImplementedError("Non-relative patterns are unsupported") + selector = _make_selector(("**",) + tuple(pattern_parts)) + for p in selector.select_from(self): + yield p + + def absolute(self): + """Return an absolute version of this path. This function works + even if the path doesn't point to anything. + + No normalization is done, i.e. all '.' and '..' will be kept along. + Use resolve() to get the canonical path to a file. + """ + # XXX untested yet! + if self._closed: + self._raise_closed() + if self.is_absolute(): + return self + # FIXME this must defer to the specific flavour (and, under Windows, + # use nt._getfullpathname()) + obj = self._from_parts([os.getcwd()] + self._parts, init=False) + obj._init(template=self) + return obj + + def resolve(self): + """ + Make the path absolute, resolving all symlinks on the way and also + normalizing it (for example turning slashes into backslashes under + Windows). + """ + if self._closed: + self._raise_closed() + s = self._flavour.resolve(self) + if s is None: + # No symlink resolution => for consistency, raise an error if + # the path doesn't exist or is forbidden + self.stat() + s = str(self.absolute()) + # Now we have no symlinks in the path, it's safe to normalize it. + normed = self._flavour.pathmod.normpath(s) + obj = self._from_parts((normed,), init=False) + obj._init(template=self) + return obj + + def stat(self): + """ + Return the result of the stat() system call on this path, like + os.stat() does. + """ + return self._accessor.stat(self) + + def owner(self): + """ + Return the login name of the file owner. + """ + import pwd + return pwd.getpwuid(self.stat().st_uid).pw_name + + def group(self): + """ + Return the group name of the file gid. + """ + import grp + return grp.getgrgid(self.stat().st_gid).gr_name + + def _raw_open(self, flags, mode=0o777): + """ + Open the file pointed by this path and return a file descriptor, + as os.open() does. + """ + if self._closed: + self._raise_closed() + return self._accessor.open(self, flags, mode) + + def open(self, mode='r', buffering=-1, encoding=None, + errors=None, newline=None): + """ + Open the file pointed by this path and return a file object, as + the built-in open() function does. + """ + if self._closed: + self._raise_closed() + return io.open(str(self), mode, buffering, encoding, errors, newline, + opener=self._opener) + + def touch(self, mode=0o666, exist_ok=True): + """ + Create this file with the given access mode, if it doesn't exist. + """ + if self._closed: + self._raise_closed() + if exist_ok: + # First try to bump modification time + # Implementation note: GNU touch uses the UTIME_NOW option of + # the utimensat() / futimens() functions. + t = time.time() + try: + self._accessor.utime(self, (t, t)) + except OSError: + # Avoid exception chaining + pass + else: + return + flags = os.O_CREAT | os.O_WRONLY + if not exist_ok: + flags |= os.O_EXCL + fd = self._raw_open(flags, mode) + os.close(fd) + + def mkdir(self, mode=0o777, parents=False): + if self._closed: + self._raise_closed() + if not parents: + self._accessor.mkdir(self, mode) + else: + try: + self._accessor.mkdir(self, mode) + except OSError as e: + if e.errno != ENOENT: + raise + self.parent.mkdir(mode, True) + self._accessor.mkdir(self, mode) + + def chmod(self, mode): + """ + Change the permissions of the path, like os.chmod(). + """ + if self._closed: + self._raise_closed() + self._accessor.chmod(self, mode) + + def lchmod(self, mode): + """ + Like chmod(), except if the path points to a symlink, the symlink's + permissions are changed, rather than its target's. + """ + if self._closed: + self._raise_closed() + self._accessor.lchmod(self, mode) + + def unlink(self): + """ + Remove this file or link. + If the path is a directory, use rmdir() instead. + """ + if self._closed: + self._raise_closed() + self._accessor.unlink(self) + + def rmdir(self): + """ + Remove this directory. The directory must be empty. + """ + if self._closed: + self._raise_closed() + self._accessor.rmdir(self) + + def lstat(self): + """ + Like stat(), except if the path points to a symlink, the symlink's + status information is returned, rather than its target's. + """ + if self._closed: + self._raise_closed() + return self._accessor.lstat(self) + + def rename(self, target): + """ + Rename this path to the given path. + """ + if self._closed: + self._raise_closed() + self._accessor.rename(self, target) + + def replace(self, target): + """ + Rename this path to the given path, clobbering the existing + destination if it exists. + """ + if self._closed: + self._raise_closed() + self._accessor.replace(self, target) + + def symlink_to(self, target, target_is_directory=False): + """ + Make this path a symlink pointing to the given path. + Note the order of arguments (self, target) is the reverse of os.symlink's. + """ + if self._closed: + self._raise_closed() + self._accessor.symlink(target, self, target_is_directory) + + # Convenience functions for querying the stat results + + def exists(self): + """ + Whether this path exists. + """ + try: + self.stat() + except OSError as e: + if e.errno != ENOENT: + raise + return False + return True + + def is_dir(self): + """ + Whether this path is a directory. + """ + try: + return S_ISDIR(self.stat().st_mode) + except OSError as e: + if e.errno != ENOENT: + raise + # Path doesn't exist or is a broken symlink + # (see https://bitbucket.org/pitrou/pathlib/issue/12/) + return False + + def is_file(self): + """ + Whether this path is a regular file (also True for symlinks pointing + to regular files). + """ + try: + return S_ISREG(self.stat().st_mode) + except OSError as e: + if e.errno != ENOENT: + raise + # Path doesn't exist or is a broken symlink + # (see https://bitbucket.org/pitrou/pathlib/issue/12/) + return False + + def is_symlink(self): + """ + Whether this path is a symbolic link. + """ + try: + return S_ISLNK(self.lstat().st_mode) + except OSError as e: + if e.errno != ENOENT: + raise + # Path doesn't exist + return False + + def is_block_device(self): + """ + Whether this path is a block device. + """ + try: + return S_ISBLK(self.stat().st_mode) + except OSError as e: + if e.errno != ENOENT: + raise + # Path doesn't exist or is a broken symlink + # (see https://bitbucket.org/pitrou/pathlib/issue/12/) + return False + + def is_char_device(self): + """ + Whether this path is a character device. + """ + try: + return S_ISCHR(self.stat().st_mode) + except OSError as e: + if e.errno != ENOENT: + raise + # Path doesn't exist or is a broken symlink + # (see https://bitbucket.org/pitrou/pathlib/issue/12/) + return False + + def is_fifo(self): + """ + Whether this path is a FIFO. + """ + try: + return S_ISFIFO(self.stat().st_mode) + except OSError as e: + if e.errno != ENOENT: + raise + # Path doesn't exist or is a broken symlink + # (see https://bitbucket.org/pitrou/pathlib/issue/12/) + return False + + def is_socket(self): + """ + Whether this path is a socket. + """ + try: + return S_ISSOCK(self.stat().st_mode) + except OSError as e: + if e.errno != ENOENT: + raise + # Path doesn't exist or is a broken symlink + # (see https://bitbucket.org/pitrou/pathlib/issue/12/) + return False + + +class PosixPath(Path, PurePosixPath): + __slots__ = () + +class WindowsPath(Path, PureWindowsPath): + __slots__ = () + diff --git a/Lib/test/test_pathlib.py b/Lib/test/test_pathlib.py new file mode 100755 --- /dev/null +++ b/Lib/test/test_pathlib.py @@ -0,0 +1,1639 @@ +import collections +import io +import os +import errno +import pathlib +import pickle +import shutil +import socket +import stat +import sys +import tempfile +import unittest +from contextlib import contextmanager + +from test import support +TESTFN = support.TESTFN + +try: + import grp, pwd +except ImportError: + grp = pwd = None + + +class _BaseFlavourTest(object): + + def _check_parse_parts(self, arg, expected): + f = self.flavour.parse_parts + sep = self.flavour.sep + altsep = self.flavour.altsep + actual = f([x.replace('/', sep) for x in arg]) + self.assertEqual(actual, expected) + if altsep: + actual = f([x.replace('/', altsep) for x in arg]) + self.assertEqual(actual, expected) + + def test_parse_parts_common(self): + check = self._check_parse_parts + sep = self.flavour.sep + # Unanchored parts + check([], ('', '', [])) + check(['a'], ('', '', ['a'])) + check(['a/'], ('', '', ['a'])) + check(['a', 'b'], ('', '', ['a', 'b'])) + # Expansion + check(['a/b'], ('', '', ['a', 'b'])) + check(['a/b/'], ('', '', ['a', 'b'])) + check(['a', 'b/c', 'd'], ('', '', ['a', 'b', 'c', 'd'])) + # Collapsing and stripping excess slashes + check(['a', 'b//c', 'd'], ('', '', ['a', 'b', 'c', 'd'])) + check(['a', 'b/c/', 'd'], ('', '', ['a', 'b', 'c', 'd'])) + # Eliminating standalone dots + check(['.'], ('', '', [])) + check(['.', '.', 'b'], ('', '', ['b'])) + check(['a', '.', 'b'], ('', '', ['a', 'b'])) + check(['a', '.', '.'], ('', '', ['a'])) + # The first part is anchored + check(['/a/b'], ('', sep, [sep, 'a', 'b'])) + check(['/a', 'b'], ('', sep, [sep, 'a', 'b'])) + check(['/a/', 'b'], ('', sep, [sep, 'a', 'b'])) + # Ignoring parts before an anchored part + check(['a', '/b', 'c'], ('', sep, [sep, 'b', 'c'])) + check(['a', '/b', '/c'], ('', sep, [sep, 'c'])) + + +class PosixFlavourTest(_BaseFlavourTest, unittest.TestCase): + flavour = pathlib._posix_flavour + + def test_parse_parts(self): + check = self._check_parse_parts + # Collapsing of excess leading slashes, except for the double-slash + # special case. + check(['//a', 'b'], ('', '//', ['//', 'a', 'b'])) + check(['///a', 'b'], ('', '/', ['/', 'a', 'b'])) + check(['////a', 'b'], ('', '/', ['/', 'a', 'b'])) + # Paths which look like NT paths aren't treated specially + check(['c:a'], ('', '', ['c:a'])) + check(['c:\\a'], ('', '', ['c:\\a'])) + check(['\\a'], ('', '', ['\\a'])) + + def test_splitroot(self): + f = self.flavour.splitroot + self.assertEqual(f(''), ('', '', '')) + self.assertEqual(f('a'), ('', '', 'a')) + self.assertEqual(f('a/b'), ('', '', 'a/b')) + self.assertEqual(f('a/b/'), ('', '', 'a/b/')) + self.assertEqual(f('/a'), ('', '/', 'a')) + self.assertEqual(f('/a/b'), ('', '/', 'a/b')) + self.assertEqual(f('/a/b/'), ('', '/', 'a/b/')) + # The root is collapsed when there are redundant slashes + # except when there are exactly two leading slashes, which + # is a special case in POSIX. + self.assertEqual(f('//a'), ('', '//', 'a')) + self.assertEqual(f('///a'), ('', '/', 'a')) + self.assertEqual(f('///a/b'), ('', '/', 'a/b')) + # Paths which look like NT paths aren't treated specially + self.assertEqual(f('c:/a/b'), ('', '', 'c:/a/b')) + self.assertEqual(f('\\/a/b'), ('', '', '\\/a/b')) + self.assertEqual(f('\\a\\b'), ('', '', '\\a\\b')) + + +class NTFlavourTest(_BaseFlavourTest, unittest.TestCase): + flavour = pathlib._windows_flavour + + def test_parse_parts(self): + check = self._check_parse_parts + # First part is anchored + check(['c:'], ('c:', '', ['c:'])) + check(['c:\\'], ('c:', '\\', ['c:\\'])) + check(['\\'], ('', '\\', ['\\'])) + check(['c:a'], ('c:', '', ['c:', 'a'])) + check(['c:\\a'], ('c:', '\\', ['c:\\', 'a'])) + check(['\\a'], ('', '\\', ['\\', 'a'])) + # UNC paths + check(['\\\\a\\b'], ('\\\\a\\b', '\\', ['\\\\a\\b\\'])) + check(['\\\\a\\b\\'], ('\\\\a\\b', '\\', ['\\\\a\\b\\'])) + check(['\\\\a\\b\\c'], ('\\\\a\\b', '\\', ['\\\\a\\b\\', 'c'])) + # Second part is anchored, so that the first part is ignored + check(['a', 'Z:b', 'c'], ('Z:', '', ['Z:', 'b', 'c'])) + check(['a', 'Z:\\b', 'c'], ('Z:', '\\', ['Z:\\', 'b', 'c'])) + check(['a', '\\b', 'c'], ('', '\\', ['\\', 'b', 'c'])) + # UNC paths + check(['a', '\\\\b\\c', 'd'], ('\\\\b\\c', '\\', ['\\\\b\\c\\', 'd'])) + # Collapsing and stripping excess slashes + check(['a', 'Z:\\\\b\\\\c\\', 'd\\'], ('Z:', '\\', ['Z:\\', 'b', 'c', 'd'])) + # UNC paths + check(['a', '\\\\b\\c\\\\', 'd'], ('\\\\b\\c', '\\', ['\\\\b\\c\\', 'd'])) + # Extended paths + check(['\\\\?\\c:\\'], ('\\\\?\\c:', '\\', ['\\\\?\\c:\\'])) + check(['\\\\?\\c:\\a'], ('\\\\?\\c:', '\\', ['\\\\?\\c:\\', 'a'])) + # Extended UNC paths (format is "\\?\UNC\server\share") + check(['\\\\?\\UNC\\b\\c'], ('\\\\?\\UNC\\b\\c', '\\', ['\\\\?\\UNC\\b\\c\\'])) + check(['\\\\?\\UNC\\b\\c\\d'], ('\\\\?\\UNC\\b\\c', '\\', ['\\\\?\\UNC\\b\\c\\', 'd'])) + + def test_splitroot(self): + f = self.flavour.splitroot + self.assertEqual(f(''), ('', '', '')) + self.assertEqual(f('a'), ('', '', 'a')) + self.assertEqual(f('a\\b'), ('', '', 'a\\b')) + self.assertEqual(f('\\a'), ('', '\\', 'a')) + self.assertEqual(f('\\a\\b'), ('', '\\', 'a\\b')) + self.assertEqual(f('c:a\\b'), ('c:', '', 'a\\b')) + self.assertEqual(f('c:\\a\\b'), ('c:', '\\', 'a\\b')) + # Redundant slashes in the root are collapsed + self.assertEqual(f('\\\\a'), ('', '\\', 'a')) + self.assertEqual(f('\\\\\\a/b'), ('', '\\', 'a/b')) + self.assertEqual(f('c:\\\\a'), ('c:', '\\', 'a')) + self.assertEqual(f('c:\\\\\\a/b'), ('c:', '\\', 'a/b')) + # Valid UNC paths + self.assertEqual(f('\\\\a\\b'), ('\\\\a\\b', '\\', '')) + self.assertEqual(f('\\\\a\\b\\'), ('\\\\a\\b', '\\', '')) + self.assertEqual(f('\\\\a\\b\\c\\d'), ('\\\\a\\b', '\\', 'c\\d')) + # These are non-UNC paths (according to ntpath.py and test_ntpath) + # However, command.com says such paths are invalid, so it's + # difficult to know what the right semantics are + self.assertEqual(f('\\\\\\a\\b'), ('', '\\', 'a\\b')) + self.assertEqual(f('\\\\a'), ('', '\\', 'a')) + + +# +# Tests for the pure classes +# + +class _BasePurePathTest(object): + + # keys are canonical paths, values are list of tuples of arguments + # supposed to produce equal paths + equivalences = { + 'a/b': [ + ('a', 'b'), ('a/', 'b'), ('a', 'b/'), ('a/', 'b/'), + ('a/b/',), ('a//b',), ('a//b//',), + # empty components get removed + ('', 'a', 'b'), ('a', '', 'b'), ('a', 'b', ''), + ], + '/b/c/d': [ + ('a', '/b/c', 'd'), ('a', '///b//c', 'd/'), + ('/a', '/b/c', 'd'), + # empty components get removed + ('/', 'b', '', 'c/d'), ('/', '', 'b/c/d'), ('', '/b/c/d'), + ], + } + + def setUp(self): + p = self.cls('a') + self.flavour = p._flavour + self.sep = self.flavour.sep + self.altsep = self.flavour.altsep + + def test_constructor_common(self): + P = self.cls + p = P('a') + self.assertIsInstance(p, P) + P('a', 'b', 'c') + P('/a', 'b', 'c') + P('a/b/c') + P('/a/b/c') + self.assertEqual(P(P('a')), P('a')) + self.assertEqual(P(P('a'), 'b'), P('a/b')) + self.assertEqual(P(P('a'), P('b')), P('a/b')) + + def test_join_common(self): + P = self.cls + p = P('a/b') + pp = p.joinpath('c') + self.assertEqual(pp, P('a/b/c')) + self.assertIs(type(pp), type(p)) + pp = p.joinpath('c', 'd') + self.assertEqual(pp, P('a/b/c/d')) + pp = p.joinpath(P('c')) + self.assertEqual(pp, P('a/b/c')) + pp = p.joinpath('/c') + self.assertEqual(pp, P('/c')) + + def test_div_common(self): + # Basically the same as joinpath() + P = self.cls + p = P('a/b') + pp = p / 'c' + self.assertEqual(pp, P('a/b/c')) + self.assertIs(type(pp), type(p)) + pp = p / 'c/d' + self.assertEqual(pp, P('a/b/c/d')) + pp = p / 'c' / 'd' + self.assertEqual(pp, P('a/b/c/d')) + pp = 'c' / p / 'd' + self.assertEqual(pp, P('c/a/b/d')) + pp = p / P('c') + self.assertEqual(pp, P('a/b/c')) + pp = p/ '/c' + self.assertEqual(pp, P('/c')) + + def _check_str(self, expected, args): + p = self.cls(*args) + self.assertEqual(str(p), expected.replace('/', self.sep)) + + def test_str_common(self): + # Canonicalized paths roundtrip + for pathstr in ('a', 'a/b', 'a/b/c', '/', '/a/b', '/a/b/c'): + self._check_str(pathstr, (pathstr,)) + # Special case for the empty path + self._check_str('.', ('',)) + # Other tests for str() are in test_equivalences() + + def test_as_posix_common(self): + P = self.cls + for pathstr in ('a', 'a/b', 'a/b/c', '/', '/a/b', '/a/b/c'): + self.assertEqual(P(pathstr).as_posix(), pathstr) + # Other tests for as_posix() are in test_equivalences() + + def test_as_bytes_common(self): + sep = os.fsencode(self.sep) + P = self.cls + self.assertEqual(bytes(P('a/b')), b'a' + sep + b'b') + + def test_as_uri_common(self): + P = self.cls + with self.assertRaises(ValueError): + P('a').as_uri() + with self.assertRaises(ValueError): + P().as_uri() + + def test_repr_common(self): + for pathstr in ('a', 'a/b', 'a/b/c', '/', '/a/b', '/a/b/c'): + p = self.cls(pathstr) + clsname = p.__class__.__name__ + r = repr(p) + # The repr() is in the form ClassName("forward-slashes path") + self.assertTrue(r.startswith(clsname + '('), r) + self.assertTrue(r.endswith(')'), r) + inner = r[len(clsname) + 1 : -1] + self.assertEqual(eval(inner), p.as_posix()) + # The repr() roundtrips + q = eval(r, pathlib.__dict__) + self.assertIs(q.__class__, p.__class__) + self.assertEqual(q, p) + self.assertEqual(repr(q), r) + + def test_eq_common(self): + P = self.cls + self.assertEqual(P('a/b'), P('a/b')) + self.assertEqual(P('a/b'), P('a', 'b')) + self.assertNotEqual(P('a/b'), P('a')) + self.assertNotEqual(P('a/b'), P('/a/b')) + self.assertNotEqual(P('a/b'), P()) + self.assertNotEqual(P('/a/b'), P('/')) + self.assertNotEqual(P(), P('/')) + self.assertNotEqual(P(), "") + self.assertNotEqual(P(), {}) + self.assertNotEqual(P(), int) + + def test_match_common(self): + P = self.cls + self.assertRaises(ValueError, P('a').match, '') + self.assertRaises(ValueError, P('a').match, '.') + # Simple relative pattern + self.assertTrue(P('b.py').match('b.py')) + self.assertTrue(P('a/b.py').match('b.py')) + self.assertTrue(P('/a/b.py').match('b.py')) + self.assertFalse(P('a.py').match('b.py')) + self.assertFalse(P('b/py').match('b.py')) + self.assertFalse(P('/a.py').match('b.py')) + self.assertFalse(P('b.py/c').match('b.py')) + # Wilcard relative pattern + self.assertTrue(P('b.py').match('*.py')) + self.assertTrue(P('a/b.py').match('*.py')) + self.assertTrue(P('/a/b.py').match('*.py')) + self.assertFalse(P('b.pyc').match('*.py')) + self.assertFalse(P('b./py').match('*.py')) + self.assertFalse(P('b.py/c').match('*.py')) + # Multi-part relative pattern + self.assertTrue(P('ab/c.py').match('a*/*.py')) + self.assertTrue(P('/d/ab/c.py').match('a*/*.py')) + self.assertFalse(P('a.py').match('a*/*.py')) + self.assertFalse(P('/dab/c.py').match('a*/*.py')) + self.assertFalse(P('ab/c.py/d').match('a*/*.py')) + # Absolute pattern + self.assertTrue(P('/b.py').match('/*.py')) + self.assertFalse(P('b.py').match('/*.py')) + self.assertFalse(P('a/b.py').match('/*.py')) + self.assertFalse(P('/a/b.py').match('/*.py')) + # Multi-part absolute pattern + self.assertTrue(P('/a/b.py').match('/a/*.py')) + self.assertFalse(P('/ab.py').match('/a/*.py')) + self.assertFalse(P('/a/b/c.py').match('/a/*.py')) + + def test_ordering_common(self): + # Ordering is tuple-alike + def assertLess(a, b): + self.assertLess(a, b) + self.assertGreater(b, a) + P = self.cls + a = P('a') + b = P('a/b') + c = P('abc') + d = P('b') + assertLess(a, b) + assertLess(a, c) + assertLess(a, d) + assertLess(b, c) + assertLess(c, d) + P = self.cls + a = P('/a') + b = P('/a/b') + c = P('/abc') + d = P('/b') + assertLess(a, b) + assertLess(a, c) + assertLess(a, d) + assertLess(b, c) + assertLess(c, d) + with self.assertRaises(TypeError): + P() < {} + + def test_parts_common(self): + # `parts` returns a tuple + sep = self.sep + P = self.cls + p = P('a/b') + parts = p.parts + self.assertEqual(parts, ('a', 'b')) + # The object gets reused + self.assertIs(parts, p.parts) + # When the path is absolute, the anchor is a separate part + p = P('/a/b') + parts = p.parts + self.assertEqual(parts, (sep, 'a', 'b')) + + def test_equivalences(self): + for k, tuples in self.equivalences.items(): + canon = k.replace('/', self.sep) + posix = k.replace(self.sep, '/') + if canon != posix: + tuples = tuples + [ + tuple(part.replace('/', self.sep) for part in t) + for t in tuples + ] + tuples.append((posix, )) + pcanon = self.cls(canon) + for t in tuples: + p = self.cls(*t) + self.assertEqual(p, pcanon, "failed with args {}".format(t)) + self.assertEqual(hash(p), hash(pcanon)) + self.assertEqual(str(p), canon) + self.assertEqual(p.as_posix(), posix) + + def test_parent_common(self): + # Relative + P = self.cls + p = P('a/b/c') + self.assertEqual(p.parent, P('a/b')) + self.assertEqual(p.parent.parent, P('a')) + self.assertEqual(p.parent.parent.parent, P()) + self.assertEqual(p.parent.parent.parent.parent, P()) + # Anchored + p = P('/a/b/c') + self.assertEqual(p.parent, P('/a/b')) + self.assertEqual(p.parent.parent, P('/a')) + self.assertEqual(p.parent.parent.parent, P('/')) + self.assertEqual(p.parent.parent.parent.parent, P('/')) + + def test_parents_common(self): + # Relative + P = self.cls + p = P('a/b/c') + par = p.parents + self.assertEqual(len(par), 3) + self.assertEqual(par[0], P('a/b')) + self.assertEqual(par[1], P('a')) + self.assertEqual(par[2], P('.')) + self.assertEqual(list(par), [P('a/b'), P('a'), P('.')]) + with self.assertRaises(IndexError): + par[-1] + with self.assertRaises(IndexError): + par[3] + with self.assertRaises(TypeError): + par[0] = p + # Anchored + p = P('/a/b/c') + par = p.parents + self.assertEqual(len(par), 3) + self.assertEqual(par[0], P('/a/b')) + self.assertEqual(par[1], P('/a')) + self.assertEqual(par[2], P('/')) + self.assertEqual(list(par), [P('/a/b'), P('/a'), P('/')]) + with self.assertRaises(IndexError): + par[3] + + def test_drive_common(self): + P = self.cls + self.assertEqual(P('a/b').drive, '') + self.assertEqual(P('/a/b').drive, '') + self.assertEqual(P('').drive, '') + + def test_root_common(self): + P = self.cls + sep = self.sep + self.assertEqual(P('').root, '') + self.assertEqual(P('a/b').root, '') + self.assertEqual(P('/').root, sep) + self.assertEqual(P('/a/b').root, sep) + + def test_anchor_common(self): + P = self.cls + sep = self.sep + self.assertEqual(P('').anchor, '') + self.assertEqual(P('a/b').anchor, '') + self.assertEqual(P('/').anchor, sep) + self.assertEqual(P('/a/b').anchor, sep) + + def test_name_common(self): + P = self.cls + self.assertEqual(P('').name, '') + self.assertEqual(P('.').name, '') + self.assertEqual(P('/').name, '') + self.assertEqual(P('a/b').name, 'b') + self.assertEqual(P('/a/b').name, 'b') + self.assertEqual(P('/a/b/.').name, 'b') + self.assertEqual(P('a/b.py').name, 'b.py') + self.assertEqual(P('/a/b.py').name, 'b.py') + + def test_suffix_common(self): + P = self.cls + self.assertEqual(P('').suffix, '') + self.assertEqual(P('.').suffix, '') + self.assertEqual(P('..').suffix, '') + self.assertEqual(P('/').suffix, '') + self.assertEqual(P('a/b').suffix, '') + self.assertEqual(P('/a/b').suffix, '') + self.assertEqual(P('/a/b/.').suffix, '') + self.assertEqual(P('a/b.py').suffix, '.py') + self.assertEqual(P('/a/b.py').suffix, '.py') + self.assertEqual(P('a/.hgrc').suffix, '') + self.assertEqual(P('/a/.hgrc').suffix, '') + self.assertEqual(P('a/.hg.rc').suffix, '.rc') + self.assertEqual(P('/a/.hg.rc').suffix, '.rc') + self.assertEqual(P('a/b.tar.gz').suffix, '.gz') + self.assertEqual(P('/a/b.tar.gz').suffix, '.gz') + self.assertEqual(P('a/Some name. Ending with a dot.').suffix, '') + self.assertEqual(P('/a/Some name. Ending with a dot.').suffix, '') + + def test_suffixes_common(self): + P = self.cls + self.assertEqual(P('').suffixes, []) + self.assertEqual(P('.').suffixes, []) + self.assertEqual(P('/').suffixes, []) + self.assertEqual(P('a/b').suffixes, []) + self.assertEqual(P('/a/b').suffixes, []) + self.assertEqual(P('/a/b/.').suffixes, []) + self.assertEqual(P('a/b.py').suffixes, ['.py']) + self.assertEqual(P('/a/b.py').suffixes, ['.py']) + self.assertEqual(P('a/.hgrc').suffixes, []) + self.assertEqual(P('/a/.hgrc').suffixes, []) + self.assertEqual(P('a/.hg.rc').suffixes, ['.rc']) + self.assertEqual(P('/a/.hg.rc').suffixes, ['.rc']) + self.assertEqual(P('a/b.tar.gz').suffixes, ['.tar', '.gz']) + self.assertEqual(P('/a/b.tar.gz').suffixes, ['.tar', '.gz']) + self.assertEqual(P('a/Some name. Ending with a dot.').suffixes, []) + self.assertEqual(P('/a/Some name. Ending with a dot.').suffixes, []) + + def test_stem_common(self): + P = self.cls + self.assertEqual(P('').stem, '') + self.assertEqual(P('.').stem, '') + self.assertEqual(P('..').stem, '..') + self.assertEqual(P('/').stem, '') + self.assertEqual(P('a/b').stem, 'b') + self.assertEqual(P('a/b.py').stem, 'b') + self.assertEqual(P('a/.hgrc').stem, '.hgrc') + self.assertEqual(P('a/.hg.rc').stem, '.hg') + self.assertEqual(P('a/b.tar.gz').stem, 'b.tar') + self.assertEqual(P('a/Some name. Ending with a dot.').stem, + 'Some name. Ending with a dot.') + + def test_with_name_common(self): + P = self.cls + self.assertEqual(P('a/b').with_name('d.xml'), P('a/d.xml')) + self.assertEqual(P('/a/b').with_name('d.xml'), P('/a/d.xml')) + self.assertEqual(P('a/b.py').with_name('d.xml'), P('a/d.xml')) + self.assertEqual(P('/a/b.py').with_name('d.xml'), P('/a/d.xml')) + self.assertEqual(P('a/Dot ending.').with_name('d.xml'), P('a/d.xml')) + self.assertEqual(P('/a/Dot ending.').with_name('d.xml'), P('/a/d.xml')) + self.assertRaises(ValueError, P('').with_name, 'd.xml') + self.assertRaises(ValueError, P('.').with_name, 'd.xml') + self.assertRaises(ValueError, P('/').with_name, 'd.xml') + + def test_with_suffix_common(self): + P = self.cls + self.assertEqual(P('a/b').with_suffix('.gz'), P('a/b.gz')) + self.assertEqual(P('/a/b').with_suffix('.gz'), P('/a/b.gz')) + self.assertEqual(P('a/b.py').with_suffix('.gz'), P('a/b.gz')) + self.assertEqual(P('/a/b.py').with_suffix('.gz'), P('/a/b.gz')) + self.assertRaises(ValueError, P('').with_suffix, '.gz') + self.assertRaises(ValueError, P('.').with_suffix, '.gz') + self.assertRaises(ValueError, P('/').with_suffix, '.gz') + + def test_relative_to_common(self): + P = self.cls + p = P('a/b') + self.assertRaises(TypeError, p.relative_to) + self.assertEqual(p.relative_to(P()), P('a/b')) + self.assertEqual(p.relative_to(P('a')), P('b')) + self.assertEqual(p.relative_to(P('a/b')), P()) + # With several args + self.assertEqual(p.relative_to('a', 'b'), P()) + # Unrelated paths + self.assertRaises(ValueError, p.relative_to, P('c')) + self.assertRaises(ValueError, p.relative_to, P('a/b/c')) + self.assertRaises(ValueError, p.relative_to, P('a/c')) + self.assertRaises(ValueError, p.relative_to, P('/a')) + p = P('/a/b') + self.assertEqual(p.relative_to(P('/')), P('a/b')) + self.assertEqual(p.relative_to(P('/a')), P('b')) + self.assertEqual(p.relative_to(P('/a/b')), P()) + # Unrelated paths + self.assertRaises(ValueError, p.relative_to, P('/c')) + self.assertRaises(ValueError, p.relative_to, P('/a/b/c')) + self.assertRaises(ValueError, p.relative_to, P('/a/c')) + self.assertRaises(ValueError, p.relative_to, P()) + self.assertRaises(ValueError, p.relative_to, P('a')) + + def test_pickling_common(self): + P = self.cls + p = P('/a/b') + for proto in range(0, pickle.HIGHEST_PROTOCOL + 1): + dumped = pickle.dumps(p, proto) + pp = pickle.loads(dumped) + self.assertIs(pp.__class__, p.__class__) + self.assertEqual(pp, p) + self.assertEqual(hash(pp), hash(p)) + self.assertEqual(str(pp), str(p)) + + +class PurePosixPathTest(_BasePurePathTest, unittest.TestCase): + cls = pathlib.PurePosixPath + + def test_root(self): + P = self.cls + self.assertEqual(P('/a/b').root, '/') + self.assertEqual(P('///a/b').root, '/') + # POSIX special case for two leading slashes + self.assertEqual(P('//a/b').root, '//') + + def test_eq(self): + P = self.cls + self.assertNotEqual(P('a/b'), P('A/b')) + self.assertEqual(P('/a'), P('///a')) + self.assertNotEqual(P('/a'), P('//a')) + + def test_as_uri(self): + from urllib.parse import quote_from_bytes + P = self.cls + self.assertEqual(P('/').as_uri(), 'file:///') + self.assertEqual(P('/a/b.c').as_uri(), 'file:///a/b.c') + self.assertEqual(P('/a/b%#c').as_uri(), 'file:///a/b%25%23c') + self.assertEqual(P('/a/b\xe9').as_uri(), + 'file:///a/b' + quote_from_bytes(os.fsencode('\xe9'))) + + def test_match(self): + P = self.cls + self.assertFalse(P('A.py').match('a.PY')) + + def test_is_absolute(self): + P = self.cls + self.assertFalse(P().is_absolute()) + self.assertFalse(P('a').is_absolute()) + self.assertFalse(P('a/b/').is_absolute()) + self.assertTrue(P('/').is_absolute()) + self.assertTrue(P('/a').is_absolute()) + self.assertTrue(P('/a/b/').is_absolute()) + self.assertTrue(P('//a').is_absolute()) + self.assertTrue(P('//a/b').is_absolute()) + + def test_is_reserved(self): + P = self.cls + self.assertIs(False, P('').is_reserved()) + self.assertIs(False, P('/').is_reserved()) + self.assertIs(False, P('/foo/bar').is_reserved()) + self.assertIs(False, P('/dev/con/PRN/NUL').is_reserved()) + + def test_join(self): + P = self.cls + p = P('//a') + pp = p.joinpath('b') + self.assertEqual(pp, P('//a/b')) + pp = P('/a').joinpath('//c') + self.assertEqual(pp, P('//c')) + pp = P('//a').joinpath('/c') + self.assertEqual(pp, P('/c')) + + def test_div(self): + # Basically the same as joinpath() + P = self.cls + p = P('//a') + pp = p / 'b' + self.assertEqual(pp, P('//a/b')) + pp = P('/a') / '//c' + self.assertEqual(pp, P('//c')) + pp = P('//a') / '/c' + self.assertEqual(pp, P('/c')) + + +class PureWindowsPathTest(_BasePurePathTest, unittest.TestCase): + cls = pathlib.PureWindowsPath + + equivalences = _BasePurePathTest.equivalences.copy() + equivalences.update({ + 'c:a': [ ('c:', 'a'), ('c:', 'a/'), ('/', 'c:', 'a') ], + 'c:/a': [ + ('c:/', 'a'), ('c:', '/', 'a'), ('c:', '/a'), + ('/z', 'c:/', 'a'), ('//x/y', 'c:/', 'a'), + ], + '//a/b/': [ ('//a/b',) ], + '//a/b/c': [ + ('//a/b', 'c'), ('//a/b/', 'c'), + ], + }) + + def test_str(self): + p = self.cls('a/b/c') + self.assertEqual(str(p), 'a\\b\\c') + p = self.cls('c:/a/b/c') + self.assertEqual(str(p), 'c:\\a\\b\\c') + p = self.cls('//a/b') + self.assertEqual(str(p), '\\\\a\\b\\') + p = self.cls('//a/b/c') + self.assertEqual(str(p), '\\\\a\\b\\c') + p = self.cls('//a/b/c/d') + self.assertEqual(str(p), '\\\\a\\b\\c\\d') + + def test_eq(self): + P = self.cls + self.assertEqual(P('c:a/b'), P('c:a/b')) + self.assertEqual(P('c:a/b'), P('c:', 'a', 'b')) + self.assertNotEqual(P('c:a/b'), P('d:a/b')) + self.assertNotEqual(P('c:a/b'), P('c:/a/b')) + self.assertNotEqual(P('/a/b'), P('c:/a/b')) + # Case-insensitivity + self.assertEqual(P('a/B'), P('A/b')) + self.assertEqual(P('C:a/B'), P('c:A/b')) + self.assertEqual(P('//Some/SHARE/a/B'), P('//somE/share/A/b')) + + def test_as_uri(self): + from urllib.parse import quote_from_bytes + P = self.cls + with self.assertRaises(ValueError): + P('/a/b').as_uri() + with self.assertRaises(ValueError): + P('c:a/b').as_uri() + self.assertEqual(P('c:/').as_uri(), 'file:///c:/') + self.assertEqual(P('c:/a/b.c').as_uri(), 'file:///c:/a/b.c') + self.assertEqual(P('c:/a/b%#c').as_uri(), 'file:///c:/a/b%25%23c') + self.assertEqual(P('c:/a/b\xe9').as_uri(), 'file:///c:/a/b%C3%A9') + self.assertEqual(P('//some/share/').as_uri(), 'file://some/share/') + self.assertEqual(P('//some/share/a/b.c').as_uri(), + 'file://some/share/a/b.c') + self.assertEqual(P('//some/share/a/b%#c\xe9').as_uri(), + 'file://some/share/a/b%25%23c%C3%A9') + + def test_match_common(self): + P = self.cls + # Absolute patterns + self.assertTrue(P('c:/b.py').match('/*.py')) + self.assertTrue(P('c:/b.py').match('c:*.py')) + self.assertTrue(P('c:/b.py').match('c:/*.py')) + self.assertFalse(P('d:/b.py').match('c:/*.py')) # wrong drive + self.assertFalse(P('b.py').match('/*.py')) + self.assertFalse(P('b.py').match('c:*.py')) + self.assertFalse(P('b.py').match('c:/*.py')) + self.assertFalse(P('c:b.py').match('/*.py')) + self.assertFalse(P('c:b.py').match('c:/*.py')) + self.assertFalse(P('/b.py').match('c:*.py')) + self.assertFalse(P('/b.py').match('c:/*.py')) + # UNC patterns + self.assertTrue(P('//some/share/a.py').match('/*.py')) + self.assertTrue(P('//some/share/a.py').match('//some/share/*.py')) + self.assertFalse(P('//other/share/a.py').match('//some/share/*.py')) + self.assertFalse(P('//some/share/a/b.py').match('//some/share/*.py')) + # Case-insensitivity + self.assertTrue(P('B.py').match('b.PY')) + self.assertTrue(P('c:/a/B.Py').match('C:/A/*.pY')) + self.assertTrue(P('//Some/Share/B.Py').match('//somE/sharE/*.pY')) + + def test_ordering_common(self): + # Case-insensitivity + def assertOrderedEqual(a, b): + self.assertLessEqual(a, b) + self.assertGreaterEqual(b, a) + P = self.cls + p = P('c:A/b') + q = P('C:a/B') + assertOrderedEqual(p, q) + self.assertFalse(p < q) + self.assertFalse(p > q) + p = P('//some/Share/A/b') + q = P('//Some/SHARE/a/B') + assertOrderedEqual(p, q) + self.assertFalse(p < q) + self.assertFalse(p > q) + + def test_parts(self): + P = self.cls + p = P('c:a/b') + parts = p.parts + self.assertEqual(parts, ('c:', 'a', 'b')) + p = P('c:/a/b') + parts = p.parts + self.assertEqual(parts, ('c:\\', 'a', 'b')) + p = P('//a/b/c/d') + parts = p.parts + self.assertEqual(parts, ('\\\\a\\b\\', 'c', 'd')) + + def test_parent(self): + # Anchored + P = self.cls + p = P('z:a/b/c') + self.assertEqual(p.parent, P('z:a/b')) + self.assertEqual(p.parent.parent, P('z:a')) + self.assertEqual(p.parent.parent.parent, P('z:')) + self.assertEqual(p.parent.parent.parent.parent, P('z:')) + p = P('z:/a/b/c') + self.assertEqual(p.parent, P('z:/a/b')) + self.assertEqual(p.parent.parent, P('z:/a')) + self.assertEqual(p.parent.parent.parent, P('z:/')) + self.assertEqual(p.parent.parent.parent.parent, P('z:/')) + p = P('//a/b/c/d') + self.assertEqual(p.parent, P('//a/b/c')) + self.assertEqual(p.parent.parent, P('//a/b')) + self.assertEqual(p.parent.parent.parent, P('//a/b')) + + def test_parents(self): + # Anchored + P = self.cls + p = P('z:a/b/') + par = p.parents + self.assertEqual(len(par), 2) + self.assertEqual(par[0], P('z:a')) + self.assertEqual(par[1], P('z:')) + self.assertEqual(list(par), [P('z:a'), P('z:')]) + with self.assertRaises(IndexError): + par[2] + p = P('z:/a/b/') + par = p.parents + self.assertEqual(len(par), 2) + self.assertEqual(par[0], P('z:/a')) + self.assertEqual(par[1], P('z:/')) + self.assertEqual(list(par), [P('z:/a'), P('z:/')]) + with self.assertRaises(IndexError): + par[2] + p = P('//a/b/c/d') + par = p.parents + self.assertEqual(len(par), 2) + self.assertEqual(par[0], P('//a/b/c')) + self.assertEqual(par[1], P('//a/b')) + self.assertEqual(list(par), [P('//a/b/c'), P('//a/b')]) + with self.assertRaises(IndexError): + par[2] + + def test_drive(self): + P = self.cls + self.assertEqual(P('c:').drive, 'c:') + self.assertEqual(P('c:a/b').drive, 'c:') + self.assertEqual(P('c:/').drive, 'c:') + self.assertEqual(P('c:/a/b/').drive, 'c:') + self.assertEqual(P('//a/b').drive, '\\\\a\\b') + self.assertEqual(P('//a/b/').drive, '\\\\a\\b') + self.assertEqual(P('//a/b/c/d').drive, '\\\\a\\b') + + def test_root(self): + P = self.cls + self.assertEqual(P('c:').root, '') + self.assertEqual(P('c:a/b').root, '') + self.assertEqual(P('c:/').root, '\\') + self.assertEqual(P('c:/a/b/').root, '\\') + self.assertEqual(P('//a/b').root, '\\') + self.assertEqual(P('//a/b/').root, '\\') + self.assertEqual(P('//a/b/c/d').root, '\\') + + def test_anchor(self): + P = self.cls + self.assertEqual(P('c:').anchor, 'c:') + self.assertEqual(P('c:a/b').anchor, 'c:') + self.assertEqual(P('c:/').anchor, 'c:\\') + self.assertEqual(P('c:/a/b/').anchor, 'c:\\') + self.assertEqual(P('//a/b').anchor, '\\\\a\\b\\') + self.assertEqual(P('//a/b/').anchor, '\\\\a\\b\\') + self.assertEqual(P('//a/b/c/d').anchor, '\\\\a\\b\\') + + def test_name(self): + P = self.cls + self.assertEqual(P('c:').name, '') + self.assertEqual(P('c:/').name, '') + self.assertEqual(P('c:a/b').name, 'b') + self.assertEqual(P('c:/a/b').name, 'b') + self.assertEqual(P('c:a/b.py').name, 'b.py') + self.assertEqual(P('c:/a/b.py').name, 'b.py') + self.assertEqual(P('//My.py/Share.php').name, '') + self.assertEqual(P('//My.py/Share.php/a/b').name, 'b') + + def test_suffix(self): + P = self.cls + self.assertEqual(P('c:').suffix, '') + self.assertEqual(P('c:/').suffix, '') + self.assertEqual(P('c:a/b').suffix, '') + self.assertEqual(P('c:/a/b').suffix, '') + self.assertEqual(P('c:a/b.py').suffix, '.py') + self.assertEqual(P('c:/a/b.py').suffix, '.py') + self.assertEqual(P('c:a/.hgrc').suffix, '') + self.assertEqual(P('c:/a/.hgrc').suffix, '') + self.assertEqual(P('c:a/.hg.rc').suffix, '.rc') + self.assertEqual(P('c:/a/.hg.rc').suffix, '.rc') + self.assertEqual(P('c:a/b.tar.gz').suffix, '.gz') + self.assertEqual(P('c:/a/b.tar.gz').suffix, '.gz') + self.assertEqual(P('c:a/Some name. Ending with a dot.').suffix, '') + self.assertEqual(P('c:/a/Some name. Ending with a dot.').suffix, '') + self.assertEqual(P('//My.py/Share.php').suffix, '') + self.assertEqual(P('//My.py/Share.php/a/b').suffix, '') + + def test_suffixes(self): + P = self.cls + self.assertEqual(P('c:').suffixes, []) + self.assertEqual(P('c:/').suffixes, []) + self.assertEqual(P('c:a/b').suffixes, []) + self.assertEqual(P('c:/a/b').suffixes, []) + self.assertEqual(P('c:a/b.py').suffixes, ['.py']) + self.assertEqual(P('c:/a/b.py').suffixes, ['.py']) + self.assertEqual(P('c:a/.hgrc').suffixes, []) + self.assertEqual(P('c:/a/.hgrc').suffixes, []) + self.assertEqual(P('c:a/.hg.rc').suffixes, ['.rc']) + self.assertEqual(P('c:/a/.hg.rc').suffixes, ['.rc']) + self.assertEqual(P('c:a/b.tar.gz').suffixes, ['.tar', '.gz']) + self.assertEqual(P('c:/a/b.tar.gz').suffixes, ['.tar', '.gz']) + self.assertEqual(P('//My.py/Share.php').suffixes, []) + self.assertEqual(P('//My.py/Share.php/a/b').suffixes, []) + self.assertEqual(P('c:a/Some name. Ending with a dot.').suffixes, []) + self.assertEqual(P('c:/a/Some name. Ending with a dot.').suffixes, []) + + def test_stem(self): + P = self.cls + self.assertEqual(P('c:').stem, '') + self.assertEqual(P('c:.').stem, '') + self.assertEqual(P('c:..').stem, '..') + self.assertEqual(P('c:/').stem, '') + self.assertEqual(P('c:a/b').stem, 'b') + self.assertEqual(P('c:a/b.py').stem, 'b') + self.assertEqual(P('c:a/.hgrc').stem, '.hgrc') + self.assertEqual(P('c:a/.hg.rc').stem, '.hg') + self.assertEqual(P('c:a/b.tar.gz').stem, 'b.tar') + self.assertEqual(P('c:a/Some name. Ending with a dot.').stem, + 'Some name. Ending with a dot.') + + def test_with_name(self): + P = self.cls + self.assertEqual(P('c:a/b').with_name('d.xml'), P('c:a/d.xml')) + self.assertEqual(P('c:/a/b').with_name('d.xml'), P('c:/a/d.xml')) + self.assertEqual(P('c:a/Dot ending.').with_name('d.xml'), P('c:a/d.xml')) + self.assertEqual(P('c:/a/Dot ending.').with_name('d.xml'), P('c:/a/d.xml')) + self.assertRaises(ValueError, P('c:').with_name, 'd.xml') + self.assertRaises(ValueError, P('c:/').with_name, 'd.xml') + self.assertRaises(ValueError, P('//My/Share').with_name, 'd.xml') + + def test_with_suffix(self): + P = self.cls + self.assertEqual(P('c:a/b').with_suffix('.gz'), P('c:a/b.gz')) + self.assertEqual(P('c:/a/b').with_suffix('.gz'), P('c:/a/b.gz')) + self.assertEqual(P('c:a/b.py').with_suffix('.gz'), P('c:a/b.gz')) + self.assertEqual(P('c:/a/b.py').with_suffix('.gz'), P('c:/a/b.gz')) + self.assertRaises(ValueError, P('').with_suffix, '.gz') + self.assertRaises(ValueError, P('.').with_suffix, '.gz') + self.assertRaises(ValueError, P('/').with_suffix, '.gz') + self.assertRaises(ValueError, P('//My/Share').with_suffix, '.gz') + + def test_relative_to(self): + P = self.cls + p = P('c:a/b') + self.assertEqual(p.relative_to(P('c:')), P('a/b')) + self.assertEqual(p.relative_to(P('c:a')), P('b')) + self.assertEqual(p.relative_to(P('c:a/b')), P()) + # Unrelated paths + self.assertRaises(ValueError, p.relative_to, P()) + self.assertRaises(ValueError, p.relative_to, P('d:')) + self.assertRaises(ValueError, p.relative_to, P('a')) + self.assertRaises(ValueError, p.relative_to, P('/a')) + self.assertRaises(ValueError, p.relative_to, P('c:a/b/c')) + self.assertRaises(ValueError, p.relative_to, P('c:a/c')) + self.assertRaises(ValueError, p.relative_to, P('c:/a')) + p = P('c:/a/b') + self.assertEqual(p.relative_to(P('c:')), P('/a/b')) + self.assertEqual(p.relative_to(P('c:/')), P('a/b')) + self.assertEqual(p.relative_to(P('c:/a')), P('b')) + self.assertEqual(p.relative_to(P('c:/a/b')), P()) + # Unrelated paths + self.assertRaises(ValueError, p.relative_to, P('c:/c')) + self.assertRaises(ValueError, p.relative_to, P('c:/a/b/c')) + self.assertRaises(ValueError, p.relative_to, P('c:/a/c')) + self.assertRaises(ValueError, p.relative_to, P('c:a')) + self.assertRaises(ValueError, p.relative_to, P('d:')) + self.assertRaises(ValueError, p.relative_to, P('d:/')) + self.assertRaises(ValueError, p.relative_to, P('/a')) + self.assertRaises(ValueError, p.relative_to, P('//c/a')) + # UNC paths + p = P('//a/b/c/d') + self.assertEqual(p.relative_to(P('//a/b')), P('c/d')) + self.assertEqual(p.relative_to(P('//a/b/c')), P('d')) + self.assertEqual(p.relative_to(P('//a/b/c/d')), P()) + # Unrelated paths + self.assertRaises(ValueError, p.relative_to, P('/a/b/c')) + self.assertRaises(ValueError, p.relative_to, P('c:/a/b/c')) + self.assertRaises(ValueError, p.relative_to, P('//z/b/c')) + self.assertRaises(ValueError, p.relative_to, P('//a/z/c')) + + def test_is_absolute(self): + P = self.cls + # Under NT, only paths with both a drive and a root are absolute + self.assertFalse(P().is_absolute()) + self.assertFalse(P('a').is_absolute()) + self.assertFalse(P('a/b/').is_absolute()) + self.assertFalse(P('/').is_absolute()) + self.assertFalse(P('/a').is_absolute()) + self.assertFalse(P('/a/b/').is_absolute()) + self.assertFalse(P('c:').is_absolute()) + self.assertFalse(P('c:a').is_absolute()) + self.assertFalse(P('c:a/b/').is_absolute()) + self.assertTrue(P('c:/').is_absolute()) + self.assertTrue(P('c:/a').is_absolute()) + self.assertTrue(P('c:/a/b/').is_absolute()) + # UNC paths are absolute by definition + self.assertTrue(P('//a/b').is_absolute()) + self.assertTrue(P('//a/b/').is_absolute()) + self.assertTrue(P('//a/b/c').is_absolute()) + self.assertTrue(P('//a/b/c/d').is_absolute()) + + def test_is_reserved(self): + P = self.cls + self.assertIs(False, P('').is_reserved()) + self.assertIs(False, P('/').is_reserved()) + self.assertIs(False, P('/foo/bar').is_reserved()) + self.assertIs(True, P('con').is_reserved()) + self.assertIs(True, P('NUL').is_reserved()) + self.assertIs(True, P('NUL.txt').is_reserved()) + self.assertIs(True, P('com1').is_reserved()) + self.assertIs(True, P('com9.bar').is_reserved()) + self.assertIs(False, P('bar.com9').is_reserved()) + self.assertIs(True, P('lpt1').is_reserved()) + self.assertIs(True, P('lpt9.bar').is_reserved()) + self.assertIs(False, P('bar.lpt9').is_reserved()) + # Only the last component matters + self.assertIs(False, P('c:/NUL/con/baz').is_reserved()) + # UNC paths are never reserved + self.assertIs(False, P('//my/share/nul/con/aux').is_reserved()) + + +class PurePathTest(_BasePurePathTest, unittest.TestCase): + cls = pathlib.PurePath + + def test_concrete_class(self): + p = self.cls('a') + self.assertIs(type(p), + pathlib.PureWindowsPath if os.name == 'nt' else pathlib.PurePosixPath) + + def test_different_flavours_unequal(self): + p = pathlib.PurePosixPath('a') + q = pathlib.PureWindowsPath('a') + self.assertNotEqual(p, q) + + def test_different_flavours_unordered(self): + p = pathlib.PurePosixPath('a') + q = pathlib.PureWindowsPath('a') + with self.assertRaises(TypeError): + p < q + with self.assertRaises(TypeError): + p <= q + with self.assertRaises(TypeError): + p > q + with self.assertRaises(TypeError): + p >= q + + +# +# Tests for the concrete classes +# + +# Make sure any symbolic links in the base test path are resolved +BASE = os.path.realpath(TESTFN) +join = lambda *x: os.path.join(BASE, *x) +rel_join = lambda *x: os.path.join(TESTFN, *x) + +def symlink_skip_reason(): + if not pathlib.supports_symlinks: + return "no system support for symlinks" + try: + os.symlink(__file__, BASE) + except OSError as e: + return str(e) + else: + support.unlink(BASE) + return None + +symlink_skip_reason = symlink_skip_reason() + +only_nt = unittest.skipIf(os.name != 'nt', + 'test requires a Windows-compatible system') +only_posix = unittest.skipIf(os.name == 'nt', + 'test requires a POSIX-compatible system') +with_symlinks = unittest.skipIf(symlink_skip_reason, symlink_skip_reason) + + +@only_posix +class PosixPathAsPureTest(PurePosixPathTest): + cls = pathlib.PosixPath + +@only_nt +class WindowsPathAsPureTest(PureWindowsPathTest): + cls = pathlib.WindowsPath + + +class _BasePathTest(object): + """Tests for the FS-accessing functionalities of the Path classes.""" + + # (BASE) + # | + # |-- dirA/ + # |-- linkC -> "../dirB" + # |-- dirB/ + # | |-- fileB + # |-- linkD -> "../dirB" + # |-- dirC/ + # | |-- fileC + # | |-- fileD + # |-- fileA + # |-- linkA -> "fileA" + # |-- linkB -> "dirB" + # + + def setUp(self): + os.mkdir(BASE) + self.addCleanup(shutil.rmtree, BASE) + os.mkdir(join('dirA')) + os.mkdir(join('dirB')) + os.mkdir(join('dirC')) + os.mkdir(join('dirC', 'dirD')) + with open(join('fileA'), 'wb') as f: + f.write(b"this is file A\n") + with open(join('dirB', 'fileB'), 'wb') as f: + f.write(b"this is file B\n") + with open(join('dirC', 'fileC'), 'wb') as f: + f.write(b"this is file C\n") + with open(join('dirC', 'dirD', 'fileD'), 'wb') as f: + f.write(b"this is file D\n") + if not symlink_skip_reason: + if os.name == 'nt': + # Workaround for http://bugs.python.org/issue13772 + def dirlink(src, dest): + os.symlink(src, dest, target_is_directory=True) + else: + def dirlink(src, dest): + os.symlink(src, dest) + # Relative symlinks + os.symlink('fileA', join('linkA')) + os.symlink('non-existing', join('brokenLink')) + dirlink('dirB', join('linkB')) + dirlink(os.path.join('..', 'dirB'), join('dirA', 'linkC')) + # This one goes upwards but doesn't create a loop + dirlink(os.path.join('..', 'dirB'), join('dirB', 'linkD')) + + def assertSame(self, path_a, path_b): + self.assertTrue(os.path.samefile(str(path_a), str(path_b)), + "%r and %r don't point to the same file" % + (path_a, path_b)) + + def assertFileNotFound(self, func, *args, **kwargs): + with self.assertRaises(FileNotFoundError) as cm: + func(*args, **kwargs) + self.assertEqual(cm.exception.errno, errno.ENOENT) + + def _test_cwd(self, p): + q = self.cls(os.getcwd()) + self.assertEqual(p, q) + self.assertEqual(str(p), str(q)) + self.assertIs(type(p), type(q)) + self.assertTrue(p.is_absolute()) + + def test_cwd(self): + p = self.cls.cwd() + self._test_cwd(p) + + def test_empty_path(self): + # The empty path points to '.' + p = self.cls('') + self.assertEqual(p.stat(), os.stat('.')) + + def test_exists(self): + P = self.cls + p = P(BASE) + self.assertIs(True, p.exists()) + self.assertIs(True, (p / 'dirA').exists()) + self.assertIs(True, (p / 'fileA').exists()) + if not symlink_skip_reason: + self.assertIs(True, (p / 'linkA').exists()) + self.assertIs(True, (p / 'linkB').exists()) + self.assertIs(False, (p / 'foo').exists()) + self.assertIs(False, P('/xyzzy').exists()) + + def test_open_common(self): + p = self.cls(BASE) + with (p / 'fileA').open('r') as f: + self.assertIsInstance(f, io.TextIOBase) + self.assertEqual(f.read(), "this is file A\n") + with (p / 'fileA').open('rb') as f: + self.assertIsInstance(f, io.BufferedIOBase) + self.assertEqual(f.read().strip(), b"this is file A") + with (p / 'fileA').open('rb', buffering=0) as f: + self.assertIsInstance(f, io.RawIOBase) + self.assertEqual(f.read().strip(), b"this is file A") + + def test_iterdir(self): + P = self.cls + p = P(BASE) + it = p.iterdir() + paths = set(it) + expected = ['dirA', 'dirB', 'dirC', 'fileA'] + if not symlink_skip_reason: + expected += ['linkA', 'linkB', 'brokenLink'] + self.assertEqual(paths, { P(BASE, q) for q in expected }) + + @with_symlinks + def test_iterdir_symlink(self): + # __iter__ on a symlink to a directory + P = self.cls + p = P(BASE, 'linkB') + paths = set(p.iterdir()) + expected = { P(BASE, 'linkB', q) for q in ['fileB', 'linkD'] } + self.assertEqual(paths, expected) + + def test_iterdir_nodir(self): + # __iter__ on something that is not a directory + p = self.cls(BASE, 'fileA') + with self.assertRaises(OSError) as cm: + next(p.iterdir()) + # ENOENT or EINVAL under Windows, ENOTDIR otherwise + # (see issue #12802) + self.assertIn(cm.exception.errno, (errno.ENOTDIR, + errno.ENOENT, errno.EINVAL)) + + def test_glob_common(self): + def _check(glob, expected): + self.assertEqual(set(glob), { P(BASE, q) for q in expected }) + P = self.cls + p = P(BASE) + it = p.glob("fileA") + self.assertIsInstance(it, collections.Iterator) + _check(it, ["fileA"]) + _check(p.glob("fileB"), []) + _check(p.glob("dir*/file*"), ["dirB/fileB", "dirC/fileC"]) + if symlink_skip_reason: + _check(p.glob("*A"), ['dirA', 'fileA']) + else: + _check(p.glob("*A"), ['dirA', 'fileA', 'linkA']) + if symlink_skip_reason: + _check(p.glob("*B/*"), ['dirB/fileB']) + else: + _check(p.glob("*B/*"), ['dirB/fileB', 'dirB/linkD', + 'linkB/fileB', 'linkB/linkD']) + if symlink_skip_reason: + _check(p.glob("*/fileB"), ['dirB/fileB']) + else: + _check(p.glob("*/fileB"), ['dirB/fileB', 'linkB/fileB']) + + def test_rglob_common(self): + def _check(glob, expected): + self.assertEqual(set(glob), { P(BASE, q) for q in expected }) + P = self.cls + p = P(BASE) + it = p.rglob("fileA") + self.assertIsInstance(it, collections.Iterator) + # XXX cannot test because of symlink loops in the test setup + #_check(it, ["fileA"]) + #_check(p.rglob("fileB"), ["dirB/fileB"]) + #_check(p.rglob("*/fileA"), [""]) + #_check(p.rglob("*/fileB"), ["dirB/fileB"]) + #_check(p.rglob("file*"), ["fileA", "dirB/fileB"]) + # No symlink loops here + p = P(BASE, "dirC") + _check(p.rglob("file*"), ["dirC/fileC", "dirC/dirD/fileD"]) + _check(p.rglob("*/*"), ["dirC/dirD/fileD"]) + + def test_glob_dotdot(self): + # ".." is not special in globs + P = self.cls + p = P(BASE) + self.assertEqual(set(p.glob("..")), { P(BASE, "..") }) + self.assertEqual(set(p.glob("dirA/../file*")), { P(BASE, "dirA/../fileA") }) + self.assertEqual(set(p.glob("../xyzzy")), set()) + + def _check_resolve_relative(self, p, expected): + q = p.resolve() + self.assertEqual(q, expected) + + def _check_resolve_absolute(self, p, expected): + q = p.resolve() + self.assertEqual(q, expected) + + @with_symlinks + def test_resolve_common(self): + P = self.cls + p = P(BASE, 'foo') + with self.assertRaises(OSError) as cm: + p.resolve() + self.assertEqual(cm.exception.errno, errno.ENOENT) + # These are all relative symlinks + p = P(BASE, 'dirB', 'fileB') + self._check_resolve_relative(p, p) + p = P(BASE, 'linkA') + self._check_resolve_relative(p, P(BASE, 'fileA')) + p = P(BASE, 'dirA', 'linkC', 'fileB') + self._check_resolve_relative(p, P(BASE, 'dirB', 'fileB')) + p = P(BASE, 'dirB', 'linkD', 'fileB') + self._check_resolve_relative(p, P(BASE, 'dirB', 'fileB')) + # Now create absolute symlinks + d = tempfile.mkdtemp(suffix='-dirD') + self.addCleanup(shutil.rmtree, d) + os.symlink(os.path.join(d), join('dirA', 'linkX')) + os.symlink(join('dirB'), os.path.join(d, 'linkY')) + p = P(BASE, 'dirA', 'linkX', 'linkY', 'fileB') + self._check_resolve_absolute(p, P(BASE, 'dirB', 'fileB')) + + def test_with(self): + p = self.cls(BASE) + it = p.iterdir() + it2 = p.iterdir() + next(it2) + with p: + pass + # I/O operation on closed path + self.assertRaises(ValueError, next, it) + self.assertRaises(ValueError, next, it2) + self.assertRaises(ValueError, p.open) + self.assertRaises(ValueError, p.resolve) + self.assertRaises(ValueError, p.absolute) + self.assertRaises(ValueError, p.__enter__) + + def test_chmod(self): + p = self.cls(BASE) / 'fileA' + mode = p.stat().st_mode + # Clear writable bit + new_mode = mode & ~0o222 + p.chmod(new_mode) + self.assertEqual(p.stat().st_mode, new_mode) + # Set writable bit + new_mode = mode | 0o222 + p.chmod(new_mode) + self.assertEqual(p.stat().st_mode, new_mode) + + # XXX also need a test for lchmod + + def test_stat(self): + p = self.cls(BASE) / 'fileA' + st = p.stat() + self.assertEqual(p.stat(), st) + # Change file mode by flipping write bit + p.chmod(st.st_mode ^ 0o222) + self.addCleanup(p.chmod, st.st_mode) + self.assertNotEqual(p.stat(), st) + + @with_symlinks + def test_lstat(self): + p = self.cls(BASE)/ 'linkA' + st = p.stat() + self.assertNotEqual(st, p.lstat()) + + def test_lstat_nosymlink(self): + p = self.cls(BASE) / 'fileA' + st = p.stat() + self.assertEqual(st, p.lstat()) + + @unittest.skipUnless(pwd, "the pwd module is needed for this test") + def test_owner(self): + p = self.cls(BASE) / 'fileA' + uid = p.stat().st_uid + name = pwd.getpwuid(uid).pw_name + self.assertEqual(name, p.owner()) + + @unittest.skipUnless(grp, "the grp module is needed for this test") + def test_group(self): + p = self.cls(BASE) / 'fileA' + gid = p.stat().st_gid + name = grp.getgrgid(gid).gr_name + self.assertEqual(name, p.group()) + + def test_unlink(self): + p = self.cls(BASE) / 'fileA' + p.unlink() + self.assertFileNotFound(p.stat) + self.assertFileNotFound(p.unlink) + + def test_rmdir(self): + p = self.cls(BASE) / 'dirA' + for q in p.iterdir(): + q.unlink() + p.rmdir() + self.assertFileNotFound(p.stat) + self.assertFileNotFound(p.unlink) + + def test_rename(self): + P = self.cls(BASE) + p = P / 'fileA' + size = p.stat().st_size + # Renaming to another path + q = P / 'dirA' / 'fileAA' + p.rename(q) + self.assertEqual(q.stat().st_size, size) + self.assertFileNotFound(p.stat) + # Renaming to a str of a relative path + r = rel_join('fileAAA') + q.rename(r) + self.assertEqual(os.stat(r).st_size, size) + self.assertFileNotFound(q.stat) + + def test_replace(self): + P = self.cls(BASE) + p = P / 'fileA' + size = p.stat().st_size + # Replacing a non-existing path + q = P / 'dirA' / 'fileAA' + p.replace(q) + self.assertEqual(q.stat().st_size, size) + self.assertFileNotFound(p.stat) + # Replacing another (existing) path + r = rel_join('dirB', 'fileB') + q.replace(r) + self.assertEqual(os.stat(r).st_size, size) + self.assertFileNotFound(q.stat) + + def test_touch_common(self): + P = self.cls(BASE) + p = P / 'newfileA' + self.assertFalse(p.exists()) + p.touch() + self.assertTrue(p.exists()) + old_mtime = p.stat().st_mtime + # Rewind the mtime sufficiently far in the past to work around + # filesystem-specific timestamp granularity. + os.utime(str(p), (old_mtime - 10, old_mtime - 10)) + # The file mtime is refreshed by calling touch() again + p.touch() + self.assertGreaterEqual(p.stat().st_mtime, old_mtime) + p = P / 'newfileB' + self.assertFalse(p.exists()) + p.touch(mode=0o700, exist_ok=False) + self.assertTrue(p.exists()) + self.assertRaises(OSError, p.touch, exist_ok=False) + + def test_mkdir(self): + P = self.cls(BASE) + p = P / 'newdirA' + self.assertFalse(p.exists()) + p.mkdir() + self.assertTrue(p.exists()) + self.assertTrue(p.is_dir()) + with self.assertRaises(OSError) as cm: + p.mkdir() + self.assertEqual(cm.exception.errno, errno.EEXIST) + # XXX test `mode` arg + + def test_mkdir_parents(self): + # Creating a chain of directories + p = self.cls(BASE, 'newdirB', 'newdirC') + self.assertFalse(p.exists()) + with self.assertRaises(OSError) as cm: + p.mkdir() + self.assertEqual(cm.exception.errno, errno.ENOENT) + p.mkdir(parents=True) + self.assertTrue(p.exists()) + self.assertTrue(p.is_dir()) + with self.assertRaises(OSError) as cm: + p.mkdir(parents=True) + self.assertEqual(cm.exception.errno, errno.EEXIST) + # XXX test `mode` arg + + @with_symlinks + def test_symlink_to(self): + P = self.cls(BASE) + target = P / 'fileA' + # Symlinking a path target + link = P / 'dirA' / 'linkAA' + link.symlink_to(target) + self.assertEqual(link.stat(), target.stat()) + self.assertNotEqual(link.lstat(), target.stat()) + # Symlinking a str target + link = P / 'dirA' / 'linkAAA' + link.symlink_to(str(target)) + self.assertEqual(link.stat(), target.stat()) + self.assertNotEqual(link.lstat(), target.stat()) + self.assertFalse(link.is_dir()) + # Symlinking to a directory + target = P / 'dirB' + link = P / 'dirA' / 'linkAAAA' + link.symlink_to(target, target_is_directory=True) + self.assertEqual(link.stat(), target.stat()) + self.assertNotEqual(link.lstat(), target.stat()) + self.assertTrue(link.is_dir()) + self.assertTrue(list(link.iterdir())) + + def test_is_dir(self): + P = self.cls(BASE) + self.assertTrue((P / 'dirA').is_dir()) + self.assertFalse((P / 'fileA').is_dir()) + self.assertFalse((P / 'non-existing').is_dir()) + if not symlink_skip_reason: + self.assertFalse((P / 'linkA').is_dir()) + self.assertTrue((P / 'linkB').is_dir()) + self.assertFalse((P/ 'brokenLink').is_dir()) + + def test_is_file(self): + P = self.cls(BASE) + self.assertTrue((P / 'fileA').is_file()) + self.assertFalse((P / 'dirA').is_file()) + self.assertFalse((P / 'non-existing').is_file()) + if not symlink_skip_reason: + self.assertTrue((P / 'linkA').is_file()) + self.assertFalse((P / 'linkB').is_file()) + self.assertFalse((P/ 'brokenLink').is_file()) + + def test_is_symlink(self): + P = self.cls(BASE) + self.assertFalse((P / 'fileA').is_symlink()) + self.assertFalse((P / 'dirA').is_symlink()) + self.assertFalse((P / 'non-existing').is_symlink()) + if not symlink_skip_reason: + self.assertTrue((P / 'linkA').is_symlink()) + self.assertTrue((P / 'linkB').is_symlink()) + self.assertTrue((P/ 'brokenLink').is_symlink()) + + def test_is_fifo_false(self): + P = self.cls(BASE) + self.assertFalse((P / 'fileA').is_fifo()) + self.assertFalse((P / 'dirA').is_fifo()) + self.assertFalse((P / 'non-existing').is_fifo()) + + @unittest.skipUnless(hasattr(os, "mkfifo"), "os.mkfifo() required") + def test_is_fifo_true(self): + P = self.cls(BASE, 'myfifo') + os.mkfifo(str(P)) + self.assertTrue(P.is_fifo()) + self.assertFalse(P.is_socket()) + self.assertFalse(P.is_file()) + + def test_is_socket_false(self): + P = self.cls(BASE) + self.assertFalse((P / 'fileA').is_socket()) + self.assertFalse((P / 'dirA').is_socket()) + self.assertFalse((P / 'non-existing').is_socket()) + + @unittest.skipUnless(hasattr(socket, "AF_UNIX"), "Unix sockets required") + def test_is_socket_true(self): + P = self.cls(BASE, 'mysock') + sock = socket.socket(socket.SOCK_STREAM, socket.AF_UNIX) + self.addCleanup(sock.close) + sock.bind(str(P)) + self.assertTrue(P.is_socket()) + self.assertFalse(P.is_fifo()) + self.assertFalse(P.is_file()) + + def test_is_block_device_false(self): + P = self.cls(BASE) + self.assertFalse((P / 'fileA').is_block_device()) + self.assertFalse((P / 'dirA').is_block_device()) + self.assertFalse((P / 'non-existing').is_block_device()) + + def test_is_char_device_false(self): + P = self.cls(BASE) + self.assertFalse((P / 'fileA').is_char_device()) + self.assertFalse((P / 'dirA').is_char_device()) + self.assertFalse((P / 'non-existing').is_char_device()) + + def test_is_char_device_true(self): + # Under Unix, /dev/null should generally be a char device + P = self.cls('/dev/null') + if not P.exists(): + self.skipTest("/dev/null required") + self.assertTrue(P.is_char_device()) + self.assertFalse(P.is_block_device()) + self.assertFalse(P.is_file()) + + def test_pickling_common(self): + p = self.cls(BASE, 'fileA') + for proto in range(0, pickle.HIGHEST_PROTOCOL + 1): + dumped = pickle.dumps(p, proto) + pp = pickle.loads(dumped) + self.assertEqual(pp.stat(), p.stat()) + + def test_parts_interning(self): + P = self.cls + p = P('/usr/bin/foo') + q = P('/usr/local/bin') + # 'usr' + self.assertIs(p.parts[1], q.parts[1]) + # 'bin' + self.assertIs(p.parts[2], q.parts[3]) + + +class PathTest(_BasePathTest, unittest.TestCase): + cls = pathlib.Path + + def test_concrete_class(self): + p = self.cls('a') + self.assertIs(type(p), + pathlib.WindowsPath if os.name == 'nt' else pathlib.PosixPath) + + def test_unsupported_flavour(self): + if os.name == 'nt': + self.assertRaises(NotImplementedError, pathlib.PosixPath) + else: + self.assertRaises(NotImplementedError, pathlib.WindowsPath) + + +@only_posix +class PosixPathTest(_BasePathTest, unittest.TestCase): + cls = pathlib.PosixPath + + def _check_symlink_loop(self, *args): + path = self.cls(*args) + with self.assertRaises(RuntimeError): + print(path.resolve()) + + def test_open_mode(self): + old_mask = os.umask(0) + self.addCleanup(os.umask, old_mask) + p = self.cls(BASE) + with (p / 'new_file').open('wb'): + pass + st = os.stat(join('new_file')) + self.assertEqual(stat.S_IMODE(st.st_mode), 0o666) + os.umask(0o022) + with (p / 'other_new_file').open('wb'): + pass + st = os.stat(join('other_new_file')) + self.assertEqual(stat.S_IMODE(st.st_mode), 0o644) + + def test_touch_mode(self): + old_mask = os.umask(0) + self.addCleanup(os.umask, old_mask) + p = self.cls(BASE) + (p / 'new_file').touch() + st = os.stat(join('new_file')) + self.assertEqual(stat.S_IMODE(st.st_mode), 0o666) + os.umask(0o022) + (p / 'other_new_file').touch() + st = os.stat(join('other_new_file')) + self.assertEqual(stat.S_IMODE(st.st_mode), 0o644) + (p / 'masked_new_file').touch(mode=0o750) + st = os.stat(join('masked_new_file')) + self.assertEqual(stat.S_IMODE(st.st_mode), 0o750) + + @with_symlinks + def test_resolve_loop(self): + # Loop detection for broken symlinks under POSIX + P = self.cls + # Loops with relative symlinks + os.symlink('linkX/inside', join('linkX')) + self._check_symlink_loop(BASE, 'linkX') + os.symlink('linkY', join('linkY')) + self._check_symlink_loop(BASE, 'linkY') + os.symlink('linkZ/../linkZ', join('linkZ')) + self._check_symlink_loop(BASE, 'linkZ') + # Loops with absolute symlinks + os.symlink(join('linkU/inside'), join('linkU')) + self._check_symlink_loop(BASE, 'linkU') + os.symlink(join('linkV'), join('linkV')) + self._check_symlink_loop(BASE, 'linkV') + os.symlink(join('linkW/../linkW'), join('linkW')) + self._check_symlink_loop(BASE, 'linkW') + + def test_glob(self): + P = self.cls + p = P(BASE) + self.assertEqual(set(p.glob("FILEa")), set()) + + def test_rglob(self): + P = self.cls + p = P(BASE, "dirC") + self.assertEqual(set(p.rglob("FILEd")), set()) + + +@only_nt +class WindowsPathTest(_BasePathTest, unittest.TestCase): + cls = pathlib.WindowsPath + + def test_glob(self): + P = self.cls + p = P(BASE) + self.assertEqual(set(p.glob("FILEa")), { P(BASE, "fileA") }) + + def test_rglob(self): + P = self.cls + p = P(BASE, "dirC") + self.assertEqual(set(p.rglob("FILEd")), { P(BASE, "dirC/dirD/fileD") }) + + +if __name__ == "__main__": + unittest.main()