Python errno 模块，ENOSYS 实例源码

def rlimit(self, resource, limits=None):
            # If pid is 0 prlimit() applies to the calling process and
            # we don't want that. We should never get here though as
            # PID 0 is not supported on Linux.
            if self.pid == 0:
                raise ValueError("can't use prlimit() against PID 0 process")
            try:
                if limits is None:
                    # get
                    return cext.linux_prlimit(self.pid, resource)
                else:
                    # set
                    if len(limits) != 2:
                        raise ValueError(
                            "second argument must be a (soft, hard) tuple, "
                            "got %s" % repr(limits))
                    soft, hard = limits
                    cext.linux_prlimit(self.pid, resource, soft, hard)
            except OSError as err:
                if err.errno == errno.ENOSYS and pid_exists(self.pid):
                    # I saw this happening on Travis:
                    # https://travis-ci.org/giampaolo/psutil/jobs/51368273
                    raise ZombieProcess(self.pid, self._name, self._ppid)
                else:
                    raise

def rlimit(self, resource, limits=None):
            # If pid is 0 prlimit() applies to the calling process and
            # we don't want that. We should never get here though as
            # PID 0 is not supported on Linux.
            if self.pid == 0:
                raise ValueError("can't use prlimit() against PID 0 process")
            try:
                if limits is None:
                    # get
                    return cext.linux_prlimit(self.pid, resource)
                else:
                    # set
                    if len(limits) != 2:
                        raise ValueError(
                            "second argument must be a (soft, hard) tuple, "
                            "got %s" % repr(limits))
                    soft, hard = limits
                    cext.linux_prlimit(self.pid, resource, soft, hard)
            except OSError as err:
                if err.errno == errno.ENOSYS and pid_exists(self.pid):
                    # I saw this happening on Travis:
                    # https://travis-ci.org/giampaolo/psutil/jobs/51368273
                    raise ZombieProcess(self.pid, self._name, self._ppid)
                else:
                    raise

def test_select_module_defines_does_not_implement_poll(self):
        # This test is to make sure that if a platform defines
        # a selector as being available but does not actually
        # implement it.

        # Reset the _DEFAULT_SELECTOR value as if using for the first time.
        selectors2._DEFAULT_SELECTOR = None

        # Now we're going to patch in a bad `poll`.
        class BadPoll(object):
            def poll(self, timeout):
                raise OSError(errno.ENOSYS)

        # Remove all selectors except `select.select` and replace `select.poll`.
        patch_select_module(self, 'select', poll=BadPoll)

        selector = self.make_selector()
        self.assertIsInstance(selector, selectors2.SelectSelector)

def rlimit(self, resource, limits=None):
            # If pid is 0 prlimit() applies to the calling process and
            # we don't want that. We should never get here though as
            # PID 0 is not supported on Linux.
            if self.pid == 0:
                raise ValueError("can't use prlimit() against PID 0 process")
            try:
                if limits is None:
                    # get
                    return cext.linux_prlimit(self.pid, resource)
                else:
                    # set
                    if len(limits) != 2:
                        raise ValueError(
                            "second argument must be a (soft, hard) tuple, "
                            "got %s" % repr(limits))
                    soft, hard = limits
                    cext.linux_prlimit(self.pid, resource, soft, hard)
            except OSError as err:
                if err.errno == errno.ENOSYS and pid_exists(self.pid):
                    # I saw this happening on Travis:
                    # https://travis-ci.org/giampaolo/psutil/jobs/51368273
                    raise ZombieProcess(self.pid, self._name, self._ppid)
                else:
                    raise

def rlimit(self, resource, limits=None):
            # If pid is 0 prlimit() applies to the calling process and
            # we don't want that. We should never get here though as
            # PID 0 is not supported on Linux.
            if self.pid == 0:
                raise ValueError("can't use prlimit() against PID 0 process")
            try:
                if limits is None:
                    # get
                    return cext.linux_prlimit(self.pid, resource)
                else:
                    # set
                    if len(limits) != 2:
                        raise ValueError(
                            "second argument must be a (soft, hard) tuple, "
                            "got %s" % repr(limits))
                    soft, hard = limits
                    cext.linux_prlimit(self.pid, resource, soft, hard)
            except OSError as err:
                if err.errno == errno.ENOSYS and pid_exists(self.pid):
                    # I saw this happening on Travis:
                    # https://travis-ci.org/giampaolo/psutil/jobs/51368273
                    raise ZombieProcess(self.pid, self._name, self._ppid)
                else:
                    raise

def rlimit(self, resource, limits=None):
            # If pid is 0 prlimit() applies to the calling process and
            # we don't want that. We should never get here though as
            # PID 0 is not supported on Linux.
            if self.pid == 0:
                raise ValueError("can't use prlimit() against PID 0 process")
            try:
                if limits is None:
                    # get
                    return cext.linux_prlimit(self.pid, resource)
                else:
                    # set
                    if len(limits) != 2:
                        raise ValueError(
                            "second argument must be a (soft, hard) tuple, "
                            "got %s" % repr(limits))
                    soft, hard = limits
                    cext.linux_prlimit(self.pid, resource, soft, hard)
            except OSError as err:
                if err.errno == errno.ENOSYS and pid_exists(self.pid):
                    # I saw this happening on Travis:
                    # https://travis-ci.org/giampaolo/psutil/jobs/51368273
                    raise ZombieProcess(self.pid, self._name, self._ppid)
                else:
                    raise

def rlimit(self, resource, limits=None):
            # If pid is 0 prlimit() applies to the calling process and
            # we don't want that. We should never get here though as
            # PID 0 is not supported on Linux.
            if self.pid == 0:
                raise ValueError("can't use prlimit() against PID 0 process")
            try:
                if limits is None:
                    # get
                    return cext.linux_prlimit(self.pid, resource)
                else:
                    # set
                    if len(limits) != 2:
                        raise ValueError(
                            "second argument must be a (soft, hard) tuple, "
                            "got %s" % repr(limits))
                    soft, hard = limits
                    cext.linux_prlimit(self.pid, resource, soft, hard)
            except OSError as err:
                if err.errno == errno.ENOSYS and pid_exists(self.pid):
                    # I saw this happening on Travis:
                    # https://travis-ci.org/giampaolo/psutil/jobs/51368273
                    raise ZombieProcess(self.pid, self._name, self._ppid)
                else:
                    raise

def rlimit(self, resource, limits=None):
            # If pid is 0 prlimit() applies to the calling process and
            # we don't want that. We should never get here though as
            # PID 0 is not supported on Linux.
            if self.pid == 0:
                raise ValueError("can't use prlimit() against PID 0 process")
            try:
                if limits is None:
                    # get
                    return cext.linux_prlimit(self.pid, resource)
                else:
                    # set
                    if len(limits) != 2:
                        raise ValueError(
                            "second argument must be a (soft, hard) tuple, "
                            "got %s" % repr(limits))
                    soft, hard = limits
                    cext.linux_prlimit(self.pid, resource, soft, hard)
            except OSError as err:
                if err.errno == errno.ENOSYS and pid_exists(self.pid):
                    # I saw this happening on Travis:
                    # https://travis-ci.org/giampaolo/psutil/jobs/51368273
                    raise ZombieProcess(self.pid, self._name, self._ppid)
                else:
                    raise

def test_rmlinkError(self):
        """
        An exception raised by L{rmlink} other than C{ENOENT} is passed up
        to the caller of L{FilesystemLock.lock}.
        """
        def fakeRmlink(name):
            raise OSError(errno.ENOSYS, None)
        self.patch(lockfile, 'rmlink', fakeRmlink)

        def fakeKill(pid, signal):
            if signal != 0:
                raise OSError(errno.EPERM, None)
            if pid == 43125:
                raise OSError(errno.ESRCH, None)
        self.patch(lockfile, 'kill', fakeKill)

        lockf = self.mktemp()

        # Make it appear locked so it has to use readlink
        lockfile.symlink(str(43125), lockf)

        lock = lockfile.FilesystemLock(lockf)
        exc = self.assertRaises(OSError, lock.lock)
        self.assertEqual(exc.errno, errno.ENOSYS)
        self.assertFalse(lock.locked)

def rlimit(self, resource, limits=None):
            # If pid is 0 prlimit() applies to the calling process and
            # we don't want that. We should never get here though as
            # PID 0 is not supported on Linux.
            if self.pid == 0:
                raise ValueError("can't use prlimit() against PID 0 process")
            try:
                if limits is None:
                    # get
                    return cext.linux_prlimit(self.pid, resource)
                else:
                    # set
                    if len(limits) != 2:
                        raise ValueError(
                            "second argument must be a (soft, hard) tuple, "
                            "got %s" % repr(limits))
                    soft, hard = limits
                    cext.linux_prlimit(self.pid, resource, soft, hard)
            except OSError as err:
                if err.errno == errno.ENOSYS and pid_exists(self.pid):
                    # I saw this happening on Travis:
                    # https://travis-ci.org/giampaolo/psutil/jobs/51368273
                    raise ZombieProcess(self.pid, self._name, self._ppid)
                else:
                    raise

def _lock(self):
        """Lock the entire multistore."""
        self._thread_lock.acquire()
        try:
            self._file.open_and_lock()
        except IOError as e:
            if e.errno == errno.ENOSYS:
                logger.warn('File system does not support locking the '
                            'credentials file.')
            elif e.errno == errno.ENOLCK:
                logger.warn('File system is out of resources for writing the '
                            'credentials file (is your disk full?).')
            elif e.errno == errno.EDEADLK:
                logger.warn('Lock contention on multistore file, opening '
                            'in read-only mode.')
            elif e.errno == errno.EACCES:
                logger.warn('Cannot access credentials file.')
            else:
                raise
        if not self._file.is_locked():
            self._read_only = True
            if self._warn_on_readonly:
                logger.warn('The credentials file (%s) is not writable. '
                            'Opening in read-only mode. Any refreshed '
                            'credentials will only be '
                            'valid for this run.', self._file.filename())
        if os.path.getsize(self._file.filename()) == 0:
            logger.debug('Initializing empty multistore file')
            # The multistore is empty so write out an empty file.
            self._data = {}
            self._write()
        elif not self._read_only or self._data is None:
            # Only refresh the data if we are read/write or we haven't
            # cached the data yet. If we are readonly, we assume is isn't
            # changing out from under us and that we only have to read it
            # once. This prevents us from whacking any new access keys that
            # we have cached in memory but were unable to write out.
            self._refresh_data_cache()

def test_select_module_defines_does_not_implement_epoll(self):
        # Same as above test except with `select.epoll`.

        # Reset the _DEFAULT_SELECTOR value as if using for the first time.
        selectors2._DEFAULT_SELECTOR = None

        # Now we're going to patch in a bad `epoll`.
        def bad_epoll(*args, **kwargs):
            raise OSError(errno.ENOSYS)

        # Remove all selectors except `select.select` and replace `select.epoll`.
        patch_select_module(self, 'select', epoll=bad_epoll)

        selector = self.make_selector()
        self.assertIsInstance(selector, selectors2.SelectSelector)

def _lock(self):
        """Lock the entire multistore."""
        self._thread_lock.acquire()
        try:
            self._file.open_and_lock()
        except (IOError, OSError) as e:
            if e.errno == errno.ENOSYS:
                logger.warn('File system does not support locking the '
                            'credentials file.')
            elif e.errno == errno.ENOLCK:
                logger.warn('File system is out of resources for writing the '
                            'credentials file (is your disk full?).')
            elif e.errno == errno.EDEADLK:
                logger.warn('Lock contention on multistore file, opening '
                            'in read-only mode.')
            elif e.errno == errno.EACCES:
                logger.warn('Cannot access credentials file.')
            else:
                raise
        if not self._file.is_locked():
            self._read_only = True
            if self._warn_on_readonly:
                logger.warn('The credentials file (%s) is not writable. '
                            'Opening in read-only mode. Any refreshed '
                            'credentials will only be '
                            'valid for this run.', self._file.filename())

        if os.path.getsize(self._file.filename()) == 0:
            logger.debug('Initializing empty multistore file')
            # The multistore is empty so write out an empty file.
            self._data = {}
            self._write()
        elif not self._read_only or self._data is None:
            # Only refresh the data if we are read/write or we haven't
            # cached the data yet. If we are readonly, we assume is isn't
            # changing out from under us and that we only have to read it
            # once. This prevents us from whacking any new access keys that
            # we have cached in memory but were unable to write out.
            self._refresh_data_cache()

def _lock(self):
        """Lock the entire multistore."""
        self._thread_lock.acquire()
        try:
            self._file.open_and_lock()
        except IOError as e:
            if e.errno == errno.ENOSYS:
                logger.warn('File system does not support locking the '
                            'credentials file.')
            elif e.errno == errno.ENOLCK:
                logger.warn('File system is out of resources for writing the '
                            'credentials file (is your disk full?).')
            else:
                raise
        if not self._file.is_locked():
            self._read_only = True
            if self._warn_on_readonly:
                logger.warn('The credentials file (%s) is not writable. '
                            'Opening in read-only mode. Any refreshed '
                            'credentials will only be '
                            'valid for this run.', self._file.filename())
        if os.path.getsize(self._file.filename()) == 0:
            logger.debug('Initializing empty multistore file')
            # The multistore is empty so write out an empty file.
            self._data = {}
            self._write()
        elif not self._read_only or self._data is None:
            # Only refresh the data if we are read/write or we haven't
            # cached the data yet. If we are readonly, we assume is isn't
            # changing out from under us and that we only have to read it
            # once. This prevents us from whacking any new access keys that
            # we have cached in memory but were unable to write out.
            self._refresh_data_cache()

def _lock(self):
        """Lock the entire multistore."""
        self._thread_lock.acquire()
        try:
            self._file.open_and_lock()
        except IOError as e:
            if e.errno == errno.ENOSYS:
                logger.warn('File system does not support locking the '
                            'credentials file.')
            elif e.errno == errno.ENOLCK:
                logger.warn('File system is out of resources for writing the '
                            'credentials file (is your disk full?).')
            else:
                raise
        if not self._file.is_locked():
            self._read_only = True
            if self._warn_on_readonly:
                logger.warn('The credentials file (%s) is not writable. '
                            'Opening in read-only mode. Any refreshed '
                            'credentials will only be '
                            'valid for this run.', self._file.filename())
        if os.path.getsize(self._file.filename()) == 0:
            logger.debug('Initializing empty multistore file')
            # The multistore is empty so write out an empty file.
            self._data = {}
            self._write()
        elif not self._read_only or self._data is None:
            # Only refresh the data if we are read/write or we haven't
            # cached the data yet. If we are readonly, we assume is isn't
            # changing out from under us and that we only have to read it
            # once. This prevents us from whacking any new access keys that
            # we have cached in memory but were unable to write out.
            self._refresh_data_cache()

def _lock(self):
        """Lock the entire multistore."""
        self._thread_lock.acquire()
        try:
            self._file.open_and_lock()
        except (IOError, OSError) as e:
            if e.errno == errno.ENOSYS:
                logger.warn('File system does not support locking the '
                            'credentials file.')
            elif e.errno == errno.ENOLCK:
                logger.warn('File system is out of resources for writing the '
                            'credentials file (is your disk full?).')
            elif e.errno == errno.EDEADLK:
                logger.warn('Lock contention on multistore file, opening '
                            'in read-only mode.')
            elif e.errno == errno.EACCES:
                logger.warn('Cannot access credentials file.')
            else:
                raise
        if not self._file.is_locked():
            self._read_only = True
            if self._warn_on_readonly:
                logger.warn('The credentials file (%s) is not writable. '
                            'Opening in read-only mode. Any refreshed '
                            'credentials will only be '
                            'valid for this run.', self._file.filename())

        if os.path.getsize(self._file.filename()) == 0:
            logger.debug('Initializing empty multistore file')
            # The multistore is empty so write out an empty file.
            self._data = {}
            self._write()
        elif not self._read_only or self._data is None:
            # Only refresh the data if we are read/write or we haven't
            # cached the data yet. If we are readonly, we assume is isn't
            # changing out from under us and that we only have to read it
            # once. This prevents us from whacking any new access keys that
            # we have cached in memory but were unable to write out.
            self._refresh_data_cache()

def test_statvfs_result_pickle(self):
        try:
            result = os.statvfs(self.fname)
        except OSError as e:
            # On AtheOS, glibc always returns ENOSYS
            if e.errno == errno.ENOSYS:
                self.skipTest('os.statvfs() failed with ENOSYS')

        for proto in range(pickle.HIGHEST_PROTOCOL + 1):
            p = pickle.dumps(result, proto)
            self.assertIn(b'statvfs_result', p)
            if proto < 4:
                self.assertIn(b'cos\nstatvfs_result\n', p)
            unpickled = pickle.loads(p)
            self.assertEqual(result, unpickled)

def _lock(self):
        """Lock the entire multistore."""
        self._thread_lock.acquire()
        try:
            self._file.open_and_lock()
        except (IOError, OSError) as e:
            if e.errno == errno.ENOSYS:
                logger.warn('File system does not support locking the '
                            'credentials file.')
            elif e.errno == errno.ENOLCK:
                logger.warn('File system is out of resources for writing the '
                            'credentials file (is your disk full?).')
            elif e.errno == errno.EDEADLK:
                logger.warn('Lock contention on multistore file, opening '
                            'in read-only mode.')
            elif e.errno == errno.EACCES:
                logger.warn('Cannot access credentials file.')
            else:
                raise
        if not self._file.is_locked():
            self._read_only = True
            if self._warn_on_readonly:
                logger.warn('The credentials file (%s) is not writable. '
                            'Opening in read-only mode. Any refreshed '
                            'credentials will only be '
                            'valid for this run.', self._file.filename())

        if os.path.getsize(self._file.filename()) == 0:
            logger.debug('Initializing empty multistore file')
            # The multistore is empty so write out an empty file.
            self._data = {}
            self._write()
        elif not self._read_only or self._data is None:
            # Only refresh the data if we are read/write or we haven't
            # cached the data yet. If we are readonly, we assume is isn't
            # changing out from under us and that we only have to read it
            # once. This prevents us from whacking any new access keys that
            # we have cached in memory but were unable to write out.
            self._refresh_data_cache()

def _lock(self):
        """Lock the entire multistore."""
        self._thread_lock.acquire()
        try:
            self._file.open_and_lock()
        except (IOError, OSError) as e:
            if e.errno == errno.ENOSYS:
                logger.warn('File system does not support locking the '
                            'credentials file.')
            elif e.errno == errno.ENOLCK:
                logger.warn('File system is out of resources for writing the '
                            'credentials file (is your disk full?).')
            elif e.errno == errno.EDEADLK:
                logger.warn('Lock contention on multistore file, opening '
                            'in read-only mode.')
            elif e.errno == errno.EACCES:
                logger.warn('Cannot access credentials file.')
            else:
                raise
        if not self._file.is_locked():
            self._read_only = True
            if self._warn_on_readonly:
                logger.warn('The credentials file (%s) is not writable. '
                            'Opening in read-only mode. Any refreshed '
                            'credentials will only be '
                            'valid for this run.', self._file.filename())

        if os.path.getsize(self._file.filename()) == 0:
            logger.debug('Initializing empty multistore file')
            # The multistore is empty so write out an empty file.
            self._data = {}
            self._write()
        elif not self._read_only or self._data is None:
            # Only refresh the data if we are read/write or we haven't
            # cached the data yet. If we are readonly, we assume is isn't
            # changing out from under us and that we only have to read it
            # once. This prevents us from whacking any new access keys that
            # we have cached in memory but were unable to write out.
            self._refresh_data_cache()

def _lock(self):
        """Lock the entire multistore."""
        self._thread_lock.acquire()
        try:
            self._file.open_and_lock()
        except (IOError, OSError) as e:
            if e.errno == errno.ENOSYS:
                logger.warn('File system does not support locking the '
                            'credentials file.')
            elif e.errno == errno.ENOLCK:
                logger.warn('File system is out of resources for writing the '
                            'credentials file (is your disk full?).')
            elif e.errno == errno.EDEADLK:
                logger.warn('Lock contention on multistore file, opening '
                            'in read-only mode.')
            elif e.errno == errno.EACCES:
                logger.warn('Cannot access credentials file.')
            else:
                raise
        if not self._file.is_locked():
            self._read_only = True
            if self._warn_on_readonly:
                logger.warn('The credentials file (%s) is not writable. '
                            'Opening in read-only mode. Any refreshed '
                            'credentials will only be '
                            'valid for this run.', self._file.filename())

        if os.path.getsize(self._file.filename()) == 0:
            logger.debug('Initializing empty multistore file')
            # The multistore is empty so write out an empty file.
            self._data = {}
            self._write()
        elif not self._read_only or self._data is None:
            # Only refresh the data if we are read/write or we haven't
            # cached the data yet. If we are readonly, we assume is isn't
            # changing out from under us and that we only have to read it
            # once. This prevents us from whacking any new access keys that
            # we have cached in memory but were unable to write out.
            self._refresh_data_cache()

def os_error_not_implemented(*unused_args, **unused_kwargs):
  raise OSError(errno.ENOSYS, 'Function not implemented')

def test_os_error_not_implemented(self):
    with self.assertRaises(OSError) as cm:
      stubs.os_error_not_implemented()
    self.mox.VerifyAll()
    e = cm.exception
    self.assertEqual(errno.ENOSYS, e.errno)
    self.assertEqual('Function not implemented', e.strerror)
    self.assertIsNone(e.filename)

def _lock(self):
    """Lock the entire multistore."""
    self._thread_lock.acquire()
    try:
      self._file.open_and_lock()
    except IOError as e:
      if e.errno == errno.ENOSYS:
        logger.warn('File system does not support locking the credentials '
                    'file.')
      elif e.errno ==  errno.ENOLCK:
        logger.warn('File system is out of resources for writing the '
                    'credentials file (is your disk full?).')
      else:
        raise
    if not self._file.is_locked():
      self._read_only = True
      if self._warn_on_readonly:
        logger.warn('The credentials file (%s) is not writable. Opening in '
                    'read-only mode. Any refreshed credentials will only be '
                    'valid for this run.', self._file.filename())
    if os.path.getsize(self._file.filename()) == 0:
      logger.debug('Initializing empty multistore file')
      # The multistore is empty so write out an empty file.
      self._data = {}
      self._write()
    elif not self._read_only or self._data is None:
      # Only refresh the data if we are read/write or we haven't
      # cached the data yet. If we are readonly, we assume is isn't
      # changing out from under us and that we only have to read it
      # once. This prevents us from whacking any new access keys that
      # we have cached in memory but were unable to write out.
      self._refresh_data_cache()

def open(self, mode):
        """
        If the mode is r+, return a L{_ReadFile} with the contents given to
        this path's initializer.

        @raise OSError: If the mode is unsupported.

        @return: A L{_ReadFile} instance
        """
        if mode == "rb+":
            return _ReadFile(self.contents)
        raise OSError(ENOSYS, "Function not implemented")

def test_symlinkError(self):
        """
        An exception raised by C{symlink} other than C{EEXIST} is passed up to
        the caller of L{FilesystemLock.lock}.
        """
        self._symlinkErrorTest(errno.ENOSYS)

def test_readlinkError(self):
        """
        An exception raised by C{readlink} other than C{ENOENT} is passed up to
        the caller of L{FilesystemLock.lock}.
        """
        self._readlinkErrorTest(OSError, errno.ENOSYS)
        self._readlinkErrorTest(IOError, errno.ENOSYS)

def test_readlinkErrorPOSIX(self):
        """
        Any L{IOError} raised by C{readlink} on a POSIX platform passed to the
        caller of L{FilesystemLock.lock}.

        On POSIX, unlike on Windows, these are unexpected errors which cannot
        be handled by L{FilesystemLock}.
        """
        self._readlinkErrorTest(IOError, errno.ENOSYS)
        self._readlinkErrorTest(IOError, errno.EACCES)

def test_statvfs_result_pickle(self):
        try:
            result = os.statvfs(self.fname)
        except OSError as e:
            # On AtheOS, glibc always returns ENOSYS
            if e.errno == errno.ENOSYS:
                self.skipTest('os.statvfs() failed with ENOSYS')

        p = pickle.dumps(result)
        self.assertIn(b'\x03cos\nstatvfs_result\n', p)
        unpickled = pickle.loads(p)
        self.assertEqual(result, unpickled)

def _lock(self):
        """Lock the entire multistore."""
        self._thread_lock.acquire()
        try:
            self._file.open_and_lock()
        except IOError as e:
            if e.errno == errno.ENOSYS:
                logger.warn('File system does not support locking the '
                            'credentials file.')
            elif e.errno == errno.ENOLCK:
                logger.warn('File system is out of resources for writing the '
                            'credentials file (is your disk full?).')
            elif e.errno == errno.EDEADLK:
                logger.warn('Lock contention on multistore file, opening '
                            'in read-only mode.')
            else:
                raise
        if not self._file.is_locked():
            self._read_only = True
            if self._warn_on_readonly:
                logger.warn('The credentials file (%s) is not writable. '
                            'Opening in read-only mode. Any refreshed '
                            'credentials will only be '
                            'valid for this run.', self._file.filename())
        if os.path.getsize(self._file.filename()) == 0:
            logger.debug('Initializing empty multistore file')
            # The multistore is empty so write out an empty file.
            self._data = {}
            self._write()
        elif not self._read_only or self._data is None:
            # Only refresh the data if we are read/write or we haven't
            # cached the data yet. If we are readonly, we assume is isn't
            # changing out from under us and that we only have to read it
            # once. This prevents us from whacking any new access keys that
            # we have cached in memory but were unable to write out.
            self._refresh_data_cache()

def statfs(self):
        tracing.trace('called')

        client_name = self.obnam.app.settings['client-name']

        total_data = sum(
            self.obnam.repo.get_generation_key(
                gen, obnamlib.REPO_GENERATION_TOTAL_DATA)
            for gen in self.obnam.repo.get_client_generation_ids(client_name))

        files = sum(
            self.obnam.repo.get_generation_key(
                gen, obnamlib.REPO_GENERATION_FILE_COUNT)
            for gen in self.obnam.repo.get_client_generation_ids(client_name))

        stv = fuse.StatVfs()
        stv.f_bsize = 65536
        stv.f_frsize = 0
        stv.f_blocks = total_data / 65536
        stv.f_bfree = 0
        stv.f_bavail = 0
        stv.f_files = files
        stv.f_ffree = 0
        stv.f_favail = 0
        stv.f_flag = 0
        stv.f_namemax = 255
        # raise OSError(errno.ENOSYS, 'Unimplemented')
        return stv

def _lock(self):
        """Lock the entire multistore."""
        self._thread_lock.acquire()
        try:
            self._file.open_and_lock()
        except IOError as e:
            if e.errno == errno.ENOSYS:
                logger.warn('File system does not support locking the '
                            'credentials file.')
            elif e.errno == errno.ENOLCK:
                logger.warn('File system is out of resources for writing the '
                            'credentials file (is your disk full?).')
            elif e.errno == errno.EDEADLK:
                logger.warn('Lock contention on multistore file, opening '
                            'in read-only mode.')
            else:
                raise
        if not self._file.is_locked():
            self._read_only = True
            if self._warn_on_readonly:
                logger.warn('The credentials file (%s) is not writable. '
                            'Opening in read-only mode. Any refreshed '
                            'credentials will only be '
                            'valid for this run.', self._file.filename())
        if os.path.getsize(self._file.filename()) == 0:
            logger.debug('Initializing empty multistore file')
            # The multistore is empty so write out an empty file.
            self._data = {}
            self._write()
        elif not self._read_only or self._data is None:
            # Only refresh the data if we are read/write or we haven't
            # cached the data yet. If we are readonly, we assume is isn't
            # changing out from under us and that we only have to read it
            # once. This prevents us from whacking any new access keys that
            # we have cached in memory but were unable to write out.
            self._refresh_data_cache()

def __call__(self, books_needed):
        '''Look up the appropriate routine or throw an error'''
        self.LCEobj.errno = errno.ENOSYS
        try:
            policy_func = self.__class__.__dict__['_policy_' + self.name]
            return policy_func(self, books_needed)
        except KeyError as e:
            # AssertionError is a "gentler" reporting path back to user
            raise AssertionError('"%s" is not implemented' % self.name)

###########################################################################
# This is NOT for testing, just a quick entry without the full Librarian.

def read(self, shelf, length, offset, fd):
        raise TmfsOSError(errno.ENOSYS)

def write(self, shelf, buf, offset, fd):
        raise TmfsOSError(errno.ENOSYS)

def setxattr(self, path, xattr, valbytes, flags, position=0):
        # flags from linux/xattr.h: XATTR_CREATE = 1, XATTR_REPLACE = 2
        if flags or position:
            raise TmfsOSError(errno.ENOSYS)     # haven't actually seen it yet

        # 'Extend' user.xxxx syntax and screen for it here
        elems = xattr.split('.')
        if elems[0] != 'user' or len(elems) < 2:
            raise TmfsOSError(errno.EINVAL)

        # Don't forget the setfattr command, and the shell it runs in, does
        # things to a "numeric" argument.  setfattr processes a leading
        # 0x and does a byte-by-byte conversion, yielding a byte array.
        # It needs pairs of digits and can be of arbitrary length.  Any
        # other argument ends up here as a pure string (well, byte array).
        try:
            value = valbytes.decode()
        except ValueError as e:
            # http://stackoverflow.com/questions/606191/convert-bytes-to-a-python-string
            value = valbytes.decode('cp437')

        rsp = self.librarian(
                self.lcp('set_xattr', path=path,
                         xattr=xattr, value=value))
        if rsp is not None:  # unexpected
            raise TmfsOSError(errno.ENOTTY)

def flush(self, path, fh):
        '''May be called zero, one, or more times per shelf open.  It's a
           chance to report delayed errors, not a syscall passthru.'''
        return 0

    # @prentry
    # def opendir(self, path, *args, **kwargs):
        # raise TmfsOSError(errno.ENOSYS)

def fsync(self, path, datasync, fh):
        raise TmfsOSError(errno.ENOSYS)

def fsyncdir(self, path, datasync, fh):
        raise TmfsOSError(errno.ENOSYS)

def bmap(self, path, blocksize, blockno):
        '''Only if "target" is a filesystem on a block device.  Convert
           file-relative blockno to device-relative block.'''
        raise TmfsOSError(errno.ENOSYS)

def chmod(self, path, mode, **kwargs):
        raise TmfsOSError(errno.ENOSYS)

def link(self, target, name):
        raise TmfsOSError(errno.ENOSYS)

def test_statvfs_attributes(self):
        if not hasattr(os, "statvfs"):
            return

        try:
            result = os.statvfs(self.fname)
        except OSError as e:
            # On AtheOS, glibc always returns ENOSYS
            if e.errno == errno.ENOSYS:
                return

        # Make sure direct access works
        self.assertEqual(result.f_bfree, result[3])

        # Make sure all the attributes are there.
        members = ('bsize', 'frsize', 'blocks', 'bfree', 'bavail', 'files',
                    'ffree', 'favail', 'flag', 'namemax')
        for value, member in enumerate(members):
            self.assertEqual(getattr(result, 'f_' + member), result[value])

        # Make sure that assignment really fails
        try:
            result.f_bfree = 1
            self.fail("No exception thrown")
        except AttributeError:
            pass

        try:
            result.parrot = 1
            self.fail("No exception thrown")
        except AttributeError:
            pass

        # Use the constructor with a too-short tuple.
        try:
            result2 = os.statvfs_result((10,))
            self.fail("No exception thrown")
        except TypeError:
            pass

        # Use the constructor with a too-long tuple.
        try:
            result2 = os.statvfs_result((0,1,2,3,4,5,6,7,8,9,10,11,12,13,14))
        except TypeError:
            pass

def test_statvfs_attributes(self):
        try:
            result = os.statvfs(self.fname)
        except OSError, e:
            # On AtheOS, glibc always returns ENOSYS
            if e.errno == errno.ENOSYS:
                self.skipTest('glibc always returns ENOSYS on AtheOS')

        # Make sure direct access works
        self.assertEqual(result.f_bfree, result[3])

        # Make sure all the attributes are there.
        members = ('bsize', 'frsize', 'blocks', 'bfree', 'bavail', 'files',
                    'ffree', 'favail', 'flag', 'namemax')
        for value, member in enumerate(members):
            self.assertEqual(getattr(result, 'f_' + member), result[value])

        # Make sure that assignment really fails
        try:
            result.f_bfree = 1
            self.fail("No exception raised")
        except TypeError:
            pass

        try:
            result.parrot = 1
            self.fail("No exception raised")
        except AttributeError:
            pass

        # Use the constructor with a too-short tuple.
        try:
            result2 = os.statvfs_result((10,))
            self.fail("No exception raised")
        except TypeError:
            pass

        # Use the constructor with a too-long tuple.
        try:
            result2 = os.statvfs_result((0,1,2,3,4,5,6,7,8,9,10,11,12,13,14))
        except TypeError:
            pass

def test_statvfs_attributes(self):
        try:
            result = os.statvfs(self.fname)
        except OSError, e:
            # On AtheOS, glibc always returns ENOSYS
            if e.errno == errno.ENOSYS:
                self.skipTest('glibc always returns ENOSYS on AtheOS')

        # Make sure direct access works
        self.assertEqual(result.f_bfree, result[3])

        # Make sure all the attributes are there.
        members = ('bsize', 'frsize', 'blocks', 'bfree', 'bavail', 'files',
                    'ffree', 'favail', 'flag', 'namemax')
        for value, member in enumerate(members):
            self.assertEqual(getattr(result, 'f_' + member), result[value])

        # Make sure that assignment really fails
        try:
            result.f_bfree = 1
            self.fail("No exception raised")
        except TypeError:
            pass

        try:
            result.parrot = 1
            self.fail("No exception raised")
        except AttributeError:
            pass

        # Use the constructor with a too-short tuple.
        try:
            result2 = os.statvfs_result((10,))
            self.fail("No exception raised")
        except TypeError:
            pass

        # Use the constructor with a too-long tuple.
        try:
            result2 = os.statvfs_result((0,1,2,3,4,5,6,7,8,9,10,11,12,13,14))
        except TypeError:
            pass

def test_statvfs_attributes(self):
        if not hasattr(os, "statvfs"):
            return

        try:
            result = os.statvfs(self.fname)
        except OSError as e:
            # On AtheOS, glibc always returns ENOSYS
            if e.errno == errno.ENOSYS:
                return

        # Make sure direct access works
        self.assertEqual(result.f_bfree, result[3])

        # Make sure all the attributes are there.
        members = ('bsize', 'frsize', 'blocks', 'bfree', 'bavail', 'files',
                    'ffree', 'favail', 'flag', 'namemax')
        for value, member in enumerate(members):
            self.assertEqual(getattr(result, 'f_' + member), result[value])

        # Make sure that assignment really fails
        try:
            result.f_bfree = 1
            self.fail("No exception raised")
        except AttributeError:
            pass

        try:
            result.parrot = 1
            self.fail("No exception raised")
        except AttributeError:
            pass

        # Use the constructor with a too-short tuple.
        try:
            result2 = os.statvfs_result((10,))
            self.fail("No exception raised")
        except TypeError:
            pass

        # Use the constructor with a too-long tuple.
        try:
            result2 = os.statvfs_result((0,1,2,3,4,5,6,7,8,9,10,11,12,13,14))
        except TypeError:
            pass

def test_statvfs_attributes(self):
        try:
            result = os.statvfs(self.fname)
        except OSError, e:
            # On AtheOS, glibc always returns ENOSYS
            if e.errno == errno.ENOSYS:
                self.skipTest('glibc always returns ENOSYS on AtheOS')

        # Make sure direct access works
        self.assertEqual(result.f_bfree, result[3])

        # Make sure all the attributes are there.
        members = ('bsize', 'frsize', 'blocks', 'bfree', 'bavail', 'files',
                    'ffree', 'favail', 'flag', 'namemax')
        for value, member in enumerate(members):
            self.assertEqual(getattr(result, 'f_' + member), result[value])

        # Make sure that assignment really fails
        try:
            result.f_bfree = 1
            self.fail("No exception raised")
        except (TypeError, AttributeError):
            pass

        try:
            result.parrot = 1
            self.fail("No exception raised")
        except AttributeError:
            pass

        # Use the constructor with a too-short tuple.
        try:
            result2 = os.statvfs_result((10,))
            self.fail("No exception raised")
        except TypeError:
            pass

        # Use the constructor with a too-long tuple.
        try:
            result2 = os.statvfs_result((0,1,2,3,4,5,6,7,8,9,10,11,12,13,14))
        except TypeError:
            pass

def test_statvfs_attributes(self):
        try:
            result = os.statvfs(self.fname)
        except OSError as e:
            # On AtheOS, glibc always returns ENOSYS
            if e.errno == errno.ENOSYS:
                self.skipTest('os.statvfs() failed with ENOSYS')

        # Make sure direct access works
        self.assertEqual(result.f_bfree, result[3])

        # Make sure all the attributes are there.
        members = ('bsize', 'frsize', 'blocks', 'bfree', 'bavail', 'files',
                    'ffree', 'favail', 'flag', 'namemax')
        for value, member in enumerate(members):
            self.assertEqual(getattr(result, 'f_' + member), result[value])

        # Make sure that assignment really fails
        try:
            result.f_bfree = 1
            self.fail("No exception raised")
        except AttributeError:
            pass

        try:
            result.parrot = 1
            self.fail("No exception raised")
        except AttributeError:
            pass

        # Use the constructor with a too-short tuple.
        try:
            result2 = os.statvfs_result((10,))
            self.fail("No exception raised")
        except TypeError:
            pass

        # Use the constructor with a too-long tuple.
        try:
            result2 = os.statvfs_result((0,1,2,3,4,5,6,7,8,9,10,11,12,13,14))
        except TypeError:
            pass

def test_statvfs_attributes(self):
        if not hasattr(os, "statvfs"):
            return

        try:
            result = os.statvfs(self.fname)
        except OSError, e:
            # On AtheOS, glibc always returns ENOSYS
            if e.errno == errno.ENOSYS:
                return

        # Make sure direct access works
        self.assertEqual(result.f_bfree, result[3])

        # Make sure all the attributes are there.
        members = ('bsize', 'frsize', 'blocks', 'bfree', 'bavail', 'files',
                    'ffree', 'favail', 'flag', 'namemax')
        for value, member in enumerate(members):
            self.assertEqual(getattr(result, 'f_' + member), result[value])

        # Make sure that assignment really fails
        try:
            result.f_bfree = 1
            self.fail("No exception raised")
        except TypeError:
            pass

        try:
            result.parrot = 1
            self.fail("No exception raised")
        except AttributeError:
            pass

        # Use the constructor with a too-short tuple.
        try:
            result2 = os.statvfs_result((10,))
            self.fail("No exception raised")
        except TypeError:
            pass

        # Use the constructor with a too-long tuple.
        try:
            result2 = os.statvfs_result((0,1,2,3,4,5,6,7,8,9,10,11,12,13,14))
        except TypeError:
            pass

def test_statvfs_attributes(self):
        try:
            result = os.statvfs(self.fname)
        except OSError as e:
            # On AtheOS, glibc always returns ENOSYS
            if e.errno == errno.ENOSYS:
                self.skipTest('os.statvfs() failed with ENOSYS')

        # Make sure direct access works
        self.assertEqual(result.f_bfree, result[3])

        # Make sure all the attributes are there.
        members = ('bsize', 'frsize', 'blocks', 'bfree', 'bavail', 'files',
                    'ffree', 'favail', 'flag', 'namemax')
        for value, member in enumerate(members):
            self.assertEqual(getattr(result, 'f_' + member), result[value])

        # Make sure that assignment really fails
        try:
            result.f_bfree = 1
            self.fail("No exception raised")
        except AttributeError:
            pass

        try:
            result.parrot = 1
            self.fail("No exception raised")
        except AttributeError:
            pass

        # Use the constructor with a too-short tuple.
        try:
            result2 = os.statvfs_result((10,))
            self.fail("No exception raised")
        except TypeError:
            pass

        # Use the constructor with a too-long tuple.
        try:
            result2 = os.statvfs_result((0,1,2,3,4,5,6,7,8,9,10,11,12,13,14))
        except TypeError:
            pass

def getxattr(self, path, xattr, position=0):
        """Called with a specific namespace.name xattr.  Can return either
           a bytes array OR an int."""
        if position:
            raise TmfsOSError(errno.ENOSYS)    # never saw this in 8 months

        rsp = self.librarian(self.lcp('get_shelf', path=path))
        shelf = TMShelf(rsp)

        # Does this also need changed to support path instead of name?
        # Piggy back for queries by kernel (globals & fault handling).
        if xattr.startswith('_obtain_'):
            # this will need some work
            data = self.shadow.getxattr(shelf, xattr)
            try:
                return bytes(data.encode())
            except AttributeError as e:     # probably the "encode()"
                self._ret_is_string = False
                return bytes(data)

        # "ls" starts with simple getattr but then comes here for
        # security.selinux, system.posix_acl_access, and posix_acl_default.
        # ls -l can also do the same thing on '/'.  Save the round trips.

        # if xattr.startswith('security.') or not shelf_name:  # path == '/'
        if xattr.startswith('security.'):  # path == '/' is legal now
            return bytes(0)

        try:
            rsp = self.librarian(
                self.lcp('get_xattr', path=path, xattr=xattr))
            value = rsp['value']
            assert value is not None    # 'No such attribute'
            if isinstance(value, int):
                return value
            elif isinstance(value, str):
                # http://stackoverflow.com/questions/606191/convert-bytes-to-a-python-string
                return bytes(value.encode('cp437'))
            else:
                bytes(value.encode())
        except Exception as e:
            raise TmfsOSError(errno.ENODATA)    # syn for ENOATTR