# -*- Mode:Python; indent-tabs-mode:nil; tab-width:4; encoding:utf-8 -*-
#
# Copyright 2002 Ben Escoto <ben@emerose.org>
# Copyright 2007 Kenneth Loafman <kenneth@loafman.com>
# Copyright 2014 Aaron Whitehouse <aaron@whitehouse.kiwi.nz>
#
# This file is part of duplicity.
#
# Duplicity is free software; you can redistribute it and/or modify it
# under the terms of the GNU General Public License as published by the
# Free Software Foundation; either version 2 of the License, or (at your
# option) any later version.
#
# Duplicity is distributed in the hope that it will be useful, but
# WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
# General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with duplicity; if not, write to the Free Software Foundation,
# Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
from builtins import next
from builtins import str
from builtins import object
import os
import stat
import sys
import re
from duplicity import config
from duplicity import diffdir
from duplicity import log
from duplicity import util
from duplicity.globmatch import GlobbingError, FilePrefixError, select_fn_from_glob
from duplicity.path import * # pylint: disable=unused-wildcard-import,redefined-builtin
u"""Iterate exactly the requested files in a directory
Parses includes and excludes to yield correct files. More
documentation on what this code does can be found on the man page.
"""
[docs]class Select(object):
u"""Iterate appropriate Paths in given directory
This class acts as an iterator on account of its next() method.
Basically, it just goes through all the files in a directory in
order (depth-first) and subjects each file to a bunch of tests
(selection functions) in order. The first test that includes or
excludes the file means that the file gets included (iterated) or
excluded. The default is include, so with no tests we would just
iterate all the files in the directory in order.
The one complication to this is that sometimes we don't know
whether or not to include a directory until we examine its
contents. For instance, if we want to include all the **.py
files. If /home/ben/foo.py exists, we should also include /home
and /home/ben, but if these directories contain no **.py files,
they shouldn't be included. For this reason, a test may not
include or exclude a directory, but merely "scan" it. If later a
file in the directory gets included, so does the directory.
As mentioned above, each test takes the form of a selection
function. The selection function takes a path, and returns:
None - means the test has nothing to say about the related file
0 - the file is excluded by the test
1 - the file is included
2 - the test says the file (must be directory) should be scanned
Also, a selection function f has a variable f.exclude which should
be true if f could potentially exclude some file. This is used
to signal an error if the last function only includes, which would
be redundant and presumably isn't what the user intends.
"""
[docs] def __init__(self, path):
u"""Initializer, called with Path of root directory"""
assert isinstance(path, Path), str(path)
self.selection_functions = []
self.rootpath = path
self.prefix = self.rootpath.uc_name
self.files_from = None
def __iter__(self): # pylint: disable=non-iterator-returned
return self
def __next__(self):
return next(self.iter)
[docs] def set_iter(self):
u"""Initialize generator, prepare to iterate."""
# Externally-accessed method
self.rootpath.setdata() # this may have changed since Select init
self.iter = self.Iterate(self.rootpath)
return self
[docs] def Iterate(self, path):
u"""Return iterator yielding paths in path
This function looks a bit more complicated than it needs to be
because it avoids extra recursion (and no extra function calls
for non-directory files) while still doing the "directory
scanning" bit.
"""
# Only called by set_iter. Internal.
def error_handler(exc, path, filename): # pylint: disable=unused-argument
fullpath = os.path.join(path.name, filename)
try:
mode = os.stat(fullpath)[stat.ST_MODE]
if stat.S_ISSOCK(mode):
log.Info(_(u"Skipping socket %s") % util.fsdecode(fullpath),
log.InfoCode.skipping_socket,
util.escape(fullpath))
else:
log.Warn(_(u"Error initializing file %s") % util.fsdecode(fullpath),
log.WarningCode.cannot_iterate,
util.escape(fullpath))
except OSError:
log.Warn(_(u"Error accessing possibly locked file %s") % util.fsdecode(fullpath),
log.WarningCode.cannot_stat, util.escape(fullpath))
return None
def dir_scanner(path):
u"""Generator of files to be included under from filesystems
Yields (subpath, scan) where scan indicates that path is a
directory and should be included *if* something inside is
subsequently included (i.e. generated later)
This generator applies the configured selection functions and
so will not yield paths which are to be excluded.
If --files-from has been specified, this will yield paths from
that filelist only, subject to any selection functions.
"""
# Only called by Iterate. Internal.
# todo: get around circular dependency issue by importing here
from duplicity import robust
if self.files_from is None:
files = robust.listpath(path)
elif path.uc_name in self.files_from:
files = self.files_from[path.uc_name]
else:
return
for filename in files:
new_path = robust.check_common_error(
error_handler, Path.append, (path, filename))
if new_path:
s = self.Select(new_path)
if (new_path.type in [u"reg", u"dir"]
and not os.access(new_path.name, os.R_OK)) \
and (s == 1 or s == 2):
# Path is a file or folder that cannot be read, but
# should be included or scanned.
log.Warn(_(u"Error accessing possibly locked file %s") %
new_path.uc_name,
log.WarningCode.cannot_read,
util.escape(new_path.name))
if diffdir.stats:
diffdir.stats.Errors += 1
elif s == 1:
# Should be included
yield (new_path, False)
elif s == 2 and new_path.isdir():
# Is a directory that should be scanned
yield (new_path, True)
if not path.type:
# base doesn't exist
log.Warn(_(u"Warning: base %s doesn't exist, continuing") %
path.uc_name)
return
log.Debug(_(u"Selecting %s") % path.uc_name)
yield path
if not path.isdir():
return
dirs_to_scan = [dir_scanner(path)]
dirs_deferred = []
while dirs_to_scan:
try:
subpath, scan = next(dirs_to_scan[-1])
except StopIteration:
dirs_to_scan.pop()
if dirs_deferred:
dirs_deferred.pop()
continue
if scan:
dirs_deferred.append(subpath)
dirs_to_scan.append(dir_scanner(subpath))
else:
if dirs_deferred:
for delayed_path in dirs_deferred:
log.Log(_(u"Selecting %s") % delayed_path.uc_name, 6)
yield delayed_path
del dirs_deferred[:]
log.Debug(_(u"Selecting %s") % subpath.uc_name)
yield subpath
if subpath.isdir():
dirs_to_scan.append(dir_scanner(subpath))
[docs] def Select(self, path):
u"""Run through the selection functions and return dominant val 0/1/2"""
# Only used by diryield and tests. Internal.
log.Debug(u"Selection: examining path %s" % path.uc_name)
if not self.selection_functions:
log.Debug(u"Selection: + no selection functions found. Including")
return 1
scan_pending = False
for sf in self.selection_functions:
result = sf(path)
log.Debug(u"Selection: result: %4s from function: %s" %
(str(result), sf.name))
if result == 2:
# Selection function says that the path should be scanned for matching files, but keep going
# through the selection functions looking for a real match (0 or 1).
scan_pending = True
elif result == 0 or result == 1:
# A real match found, no need to try other functions.
break
if scan_pending and result != 1:
# A selection function returned 2 and either no real match was
# found or the highest-priority match was 0
result = 2
if result is None:
result = 1
if result == 0:
log.Debug(u"Selection: - excluding file")
elif result == 1:
log.Debug(u"Selection: + including file")
else:
assert result == 2
log.Debug(u"Selection: ? scanning directory for matches")
return result
[docs] def ParseArgs(self, argtuples, filelists):
u"""Create selection functions based on list of tuples
The tuples are created when the initial commandline arguments
are read. They have the form (option string, additional
argument) except for the filelist tuples, which should be
(option-string, (additional argument, filelist_fp)).
"""
# Sanity checks on --filter-* options for the benefit of users
if argtuples and argtuples[-1][0].startswith(u"--filter-"):
log.FatalError(_(u"""\
The last file selection option is the filter option %s, which will have no
effect as there are no subsequent file selection options. Exiting because
this probably isn't what you meant.""") %
(argtuples[-1][0],),
log.ErrorCode.trailing_filter)
f_opt = set(opt[0] for opt in argtuples if opt[0].startswith(u"--filter-"))
f_def = (u"--filter-globbing", u"--filter-strictcase")
if f_opt and all(opt in f_def for opt in f_opt):
log.FatalError(_(u"""\
Only these filter mode options were specified:
%s
Case sensitive globbing is the default behaviour and so this has no effect.
Exiting because this probably isn't what you meant.""") %
(u", ".join(f_opt),),
log.ErrorCode.redundant_filter)
# Called by commandline.py set_selection. External.
filelists_index = 0
mode = u"globbing"
no_case = False
try:
for opt, arg in argtuples:
if opt == u"--exclude":
self.add_selection_func(self.general_get_sf(arg, 0, mode, ignore_case=no_case))
elif opt == u"--exclude-if-present":
self.add_selection_func(self.present_get_sf(arg, 0), add_to_start=True)
elif opt == u"--exclude-device-files":
self.add_selection_func(self.devfiles_get_sf(), add_to_start=True)
elif opt == u"--exclude-filelist":
for sf in self.filelist_general_get_sfs(filelists[filelists_index], 0, arg, mode, no_case):
self.add_selection_func(sf)
filelists_index += 1
elif opt == u"--exclude-globbing-filelist":
# --exclude-globbing-filelist is now deprecated, but if it
# turns up then it needs to always work in globbing mode...
for sf in self.filelist_general_get_sfs(filelists[filelists_index], 0, arg, u"globbing", no_case):
self.add_selection_func(sf)
filelists_index += 1
elif opt == u"--exclude-other-filesystems":
self.add_selection_func(self.other_filesystems_get_sf(0))
elif opt == u"--exclude-regexp":
self.add_selection_func(self.regexp_get_sf(arg, 0, no_case))
elif opt == u"--exclude-older-than":
self.add_selection_func(self.exclude_older_get_sf(arg))
elif opt == u"--filter-strictcase":
no_case = False
elif opt == u"--filter-globbing":
mode = u"globbing"
elif opt == u"--filter-ignorecase":
no_case = True
elif opt == u"--filter-literal":
mode = u"literal"
elif opt == u"--filter-regexp":
mode = u"regex"
elif opt == u"--files-from":
self.parse_files_from(filelists[filelists_index], arg)
filelists_index += 1
elif opt == u"--include":
self.add_selection_func(self.general_get_sf(arg, 1, mode, no_case))
elif opt == u"--include-filelist":
for sf in self.filelist_general_get_sfs(filelists[filelists_index], 1, arg, mode, no_case):
self.add_selection_func(sf)
filelists_index += 1
elif opt == u"--include-globbing-filelist":
# --include-globbing-filelist is now deprecated, but if it
# turns up then it needs to always work in globbing mode...
for sf in self.filelist_general_get_sfs(filelists[filelists_index], 1, arg, u"globbing", no_case):
self.add_selection_func(sf)
filelists_index += 1
elif opt == u"--include-regexp":
self.add_selection_func(self.regexp_get_sf(arg, 1, no_case))
else:
assert 0, u"Bad selection option %s" % opt
except GlobbingError as e:
self.parse_catch_error(e)
assert filelists_index == len(filelists)
self.parse_last_excludes()
[docs] def parse_catch_error(self, exc):
u"""Deal with selection error exc"""
# Internal, used by ParseArgs.
if isinstance(exc, FilePrefixError):
log.FatalError(_(u"""\
Fatal Error: The file specification
%s
cannot match any files in the base directory
%s
Useful file specifications begin with the base directory or some
pattern (such as '**') which matches the base directory.""") %
(exc, self.prefix), log.ErrorCode.file_prefix_error)
elif isinstance(exc, GlobbingError):
log.FatalError(_(u"Fatal Error while processing expression\n"
u"%s") % exc, log.ErrorCode.globbing_error)
else:
raise # pylint: disable=misplaced-bare-raise
[docs] def parse_files_from(self, filelist_fp, list_name):
u"""Loads an explicit list of files to backup from a filelist, building
a dictionary of directories and their contents which can be used later
to emulate a filesystem walk over the listed files only.
Each specified path is unwound to identify the parents folder(s) as these
are implicitly to be included.
Paths read are not to be stripped, checked for comments, etc. Every
character on each line is significant and treated as part of the path.
"""
# Internal. Used by ParseArgs.
log.Notice(_(u"Reading files to backup from list %s") % list_name)
separator = config.null_separator and u"\0" or u"\n"
filelist = {}
absolute_path = None
for line in filelist_fp.read().split(separator):
if not line: # skip blanks
continue
if line.startswith(u"/"): # no absolute paths thanks
absolute_path = line
break
while line:
dirname, basename = os.path.split(line)
path = os.path.join(self.rootpath.uc_name, dirname).rstrip(os.path.sep)
if path not in filelist:
filelist[path] = set()
if isinstance(basename, str):
filelist[path].add(util.fsencode(basename))
else:
filelist[path].add(basename)
line = dirname
if absolute_path:
log.FatalError(_(u"""\
Files-from list contains the absolute path:
%s
All paths specified in a files-from list must be given relative to the backup
source path.""") %
(absolute_path,),
log.ErrorCode.absolute_files_from)
if not filelist:
log.FatalError(_(u"""\
Files-from list specified which contains no files, the backup will be empty as
a result. Exiting as this probably isn't what you meant,"""), log.ErrorCode.empty_files_from)
self.files_from = {d: sorted(f) for d, f in filelist.items()}
[docs] def parse_last_excludes(self):
u"""Exit with error if last selection function isn't an exclude"""
# Internal. Used by ParseArgs.
if self.selection_functions and not self.selection_functions[-1].exclude:
log.FatalError(_(u"""\
Last selection expression:
%s
only specifies that files be included. Because the default is to
include all files, the expression is redundant. Exiting because this
probably isn't what you meant.""") %
(self.selection_functions[-1].name,),
log.ErrorCode.redundant_inclusion)
[docs] def add_selection_func(self, sel_func, add_to_start=None):
u"""Add another selection function at the end or beginning"""
# Internal. Used by ParseArgs.
if add_to_start:
self.selection_functions.insert(0, sel_func)
else:
self.selection_functions.append(sel_func)
[docs] def filelist_sanitise_line(self, line, include_default):
u"""
Sanitises lines of both normal and globbing filelists, returning
(line, include) and line=None if blank/comment
The aim is to parse filelists in a consistent way, prior to the
interpretation of globbing statements. The function removes
whitespace, comment lines and processes modifiers (leading +/-)
and quotes.
"""
# Internal. Used by filelist_general_get_sfs
line = line.strip()
if not line: # skip blanks
return None, include_default
if line.startswith(u"#"): # skip full-line comments
return None, include_default
include = include_default
if line.startswith(u"+ "):
# Check for "+ " or "- " syntax
include = 1
line = line[2:]
elif line.startswith(u"- "):
include = 0
line = line[2:]
if (line.startswith(u"'") and line.endswith(u"'")) or (line.startswith(u'"') and line.endswith(u'"')):
line = line[1:-1]
return line, include
[docs] def filelist_general_get_sfs(self, filelist_fp, inc_default, list_name, mode=u"globbing", ignore_case=False):
u"""Return list of selection functions by reading fileobj
filelist_fp should be an open file object
inc_default is true if this is an include list
list_name is just the name of the list, used for logging
mode indicates whether to glob, regex, or not
"""
# Internal. Used by ParseArgs.
log.Notice(_(u"Reading %s filelist %s") % (mode, list_name))
separator = config.null_separator and u"\0" or u"\n"
try:
filelist_fp.seek(0)
except:
pass
for line in filelist_fp.read().split(separator):
line, include = self.filelist_sanitise_line(line, inc_default)
if not line:
# Skip blanks and comment lines
continue
yield self.general_get_sf(line, include, mode, ignore_case)
[docs] def other_filesystems_get_sf(self, include):
u"""Return selection function matching files on other filesystems"""
# Internal. Used by ParseArgs and unit tests.
assert include == 0 or include == 1
root_devloc = self.rootpath.getdevloc()
def sel_func(path):
if path.exists() and path.getdevloc() != root_devloc:
return include
else:
return None
sel_func.exclude = not include
sel_func.name = u"Match other filesystems"
return sel_func
[docs] def regexp_get_sf(self, regexp_string, include, ignore_case=False):
u"""Return selection function given by regexp_string"""
# Internal. Used by ParseArgs and unit tests.
assert include == 0 or include == 1
flags = 0
if ignore_case:
flags = re.IGNORECASE
try:
regexp = re.compile(regexp_string, flags)
except Exception:
log.Warn(_(u"Error compiling regular expression %s") % regexp_string)
raise
def sel_func(path):
if regexp.search(path.uc_name):
return include
else:
return None
sel_func.exclude = not include
sel_func.name = u"regular expression %s %scase: %s" % \
(include and u"include" or u"exclude", ignore_case and u"no-" or u"", regexp_string)
return sel_func
[docs] def devfiles_get_sf(self):
u"""Return a selection function to exclude all dev files"""
# Internal. Used by ParseArgs.
def sel_func(path):
if path.isdev():
return 0
else:
return None
sel_func.exclude = 1
sel_func.name = u"Exclude device files"
return sel_func
[docs] def general_get_sf(self, pattern_str, include, mode=u"globbing", ignore_case=False):
u"""Return selection function given by a pattern string
The selection patterns are interpretted in accordance with the mode
argument, "globbing", "literal", or "regex".
The 'ignorecase:' prefix is a legacy feature which historically lived on
the globbing code path and was only ever documented as working for globs.
"""
# Internal. Used by ParseArgs, filelist_general_get_sfs and unit tests.
assert include == 0 or include == 1
assert isinstance(pattern_str, str)
# legacy prefix applies *only* in globbing mode
if mode == u"globbing" and pattern_str.lower().startswith(u"ignorecase:"):
pattern_str = pattern_str[len(u"ignorecase:"):]
pattern_str = util.casefold_compat(util.fsdecode(pattern_str))
ignore_case = True
if mode == u"globbing":
return self.glob_get_sf(pattern_str, include, ignore_case)
elif mode == u"literal":
return self.literal_get_sf(pattern_str, include, ignore_case)
elif mode == u"regex":
return self.regexp_get_sf(pattern_str, include, ignore_case)
else:
assert 0, u"Bad selection mode %s" % mode
[docs] def present_get_sf(self, filename, include):
u"""Return selection function given by existence of a file in a directory"""
# Internal. Used by ParseArgs.
assert include == 0 or include == 1
# todo: get around circular dependency issue by importing here
from duplicity.robust import check_common_error
def exclude_sel_func(path):
# do not follow symbolic links when checking for file existence!
if path.isdir():
def error_handler(_exc, _filename):
# Path is not read accessible
# ToDo: Ideally this error would only show if the folder
# was ultimately included by the full set of selection
# functions. Currently this will give an error for any
# locked directory within the folder being backed up.
log.Warn(_(
u"Error accessing possibly locked file %s") % path.uc_name,
log.WarningCode.cannot_read, util.escape(path.uc_name))
if diffdir.stats:
diffdir.stats.Errors += 1
return False
if check_common_error(error_handler, path.contains, [filename]):
return 0
else:
return None
if include == 0:
sel_func = exclude_sel_func
else:
log.FatalError(u"--include-if-present not implemented (would it make sense?).",
log.ErrorCode.not_implemented)
sel_func.exclude = not include
sel_func.name = u"Command-line %s filename: %s" % \
(include and u"include-if-present" or u"exclude-if-present", filename)
return sel_func
[docs] def glob_get_sf(self, glob_str, include, ignore_case=False):
u"""Return selection function based on glob_str"""
assert isinstance(glob_str, str), \
u"The glob string " + glob_str.decode(sys.getfilesystemencoding(), u"ignore") + u" is not unicode"
# Check to make sure prefix is ok, i.e. the glob string is within
# the root folder being backed up
if not select_fn_from_glob(glob_str, 1, ignore_case)(self.rootpath):
# file_prefix_selection == 1 (include) or 2 (scan)
raise FilePrefixError(glob_str)
if glob_str == u"**":
sel_func = lambda path: include
else:
sel_func = select_fn_from_glob(glob_str, include, ignore_case)
sel_func.exclude = not include
sel_func.name = u"shell glob %s %scase: %s" % \
(include and u"include" or u"exclude", ignore_case and u"no-" or u"", glob_str)
return sel_func
[docs] def literal_get_sf(self, lit_str, include, ignore_case=False):
u"""Return a selection function that matches a literal string while
still including the contents of any folders which are matched
"""
# Check to make sure prefix is ok, see also glob_get_sf()
if not self.select_fn_from_literal(lit_str, 1, ignore_case)(self.rootpath):
raise FilePrefixError(lit_str)
sel_func = self.select_fn_from_literal(lit_str, include, ignore_case)
sel_func.exclude = not include
sel_func.name = u"literal string %s %scase: %s" % \
(include and u"include" or u"exclude", ignore_case and u"no-" or u"", lit_str)
return sel_func
[docs] def exclude_older_get_sf(self, date):
u"""Return selection function based on files older than modification date """
# Internal. Used by ParseArgs.
def sel_func(path):
if not path.isreg():
return None
try:
if os.path.getmtime(path.name) < date:
return 0
except OSError as e:
pass # this is probably only on a race condition of file being deleted
return None
sel_func.exclude = True
sel_func.name = u"Select older than %s" % (date,)
return sel_func
[docs] def select_fn_from_literal(self, lit_str, include, ignore_case=False):
u"""Return a function test_fn(path) which test where a path matches a
literal string. See also select_fn_from_blog() in globmatch.py
This function is separated from literal_get_sf() so that it can be used
to test the prefix without creating a loop.
TODO: this doesn't need to be part of the Select class type, but not
sure where else to put it?
"""
if lit_str != u"/" and lit_str[-1] == u"/":
lit_str = lit_str[:-1]
if ignore_case:
lit_str = util.casefold_compat(util.fsdecode(lit_str))
def test_fn(path):
# FIXME: caller to do this once, rather than for every path?
if ignore_case:
uc_name = util.casefold_compat(util.fsdecode(path.uc_name))
else:
uc_name = path.uc_name
if uc_name == lit_str:
return include
elif uc_name.startswith(lit_str) and uc_name[len(lit_str)] == u"/":
return include
elif lit_str.startswith(uc_name) and lit_str[len(uc_name)] == u"/" and include == 1:
return 2
else:
return None
return test_fn