17816721342

Committed 18 Sep 2025 03:05AM UTC coverage: 73.39% (+0.08%) from 73.311%

Build # 17816721342

Build Type

Pull #56

github

Committed by

web-flow

Commit Message

Merge 3dfb9e71d into 50daef4a0

Pull Request Pull Request #56: Address all Ruff linter errors, either with code changes or exclusions

Run Details

139 of 277 new or added lines in 29 files covered. (50.18%)

2 existing lines in 2 files now uncovered.

7907 of 10774 relevant lines covered (73.39%)

2.93 hits per line

Source File
Press 'n' to go to next uncovered line, 'b' for previous

70.62

/src/CedarBackup3/extend/split.py

# vim: set ft=python ts=4 sw=4 expandtab:
# # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # #
#
#              C E D A R
#          S O L U T I O N S       "Software done right."
#           S O F T W A R E
#
# # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # #
#
# Copyright (c) 2007,2010,2013,2015 Kenneth J. Pronovici.
# All rights reserved.
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License,
# Version 2, as published by the Free Software Foundation.
#
# This program 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.
#
# Copies of the GNU General Public License are available from
# the Free Software Foundation website, http://www.gnu.org/.
#
# # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # #
#
# Author   : Kenneth J. Pronovici <pronovic@ieee.org>
# Language : Python 3
# Project  : Official Cedar Backup Extensions
# Purpose  : Provides an extension to split up large files in staging directories.
#
# # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # #

########################################################################
# Module documentation
########################################################################

"""
Provides an extension to split up large files in staging directories.

When this extension is executed, it will look through the configured Cedar
Backup staging directory for files exceeding a specified size limit, and split
them down into smaller files using the 'split' utility.  Any directory which
has already been split (as indicated by the ``cback.split`` file) will be
ignored.

This extension requires a new configuration section <split> and is intended
to be run immediately after the standard stage action or immediately before the
standard store action.  Aside from its own configuration, it requires the
options and staging configuration sections in the standard Cedar Backup
configuration file.

:author: Kenneth J. Pronovici <pronovic@ieee.org>
"""

########################################################################
# Imported modules
########################################################################

import logging
import os
import re
from functools import total_ordering

from CedarBackup3.actions.util import findDailyDirs, getBackupFiles, writeIndicatorFile
from CedarBackup3.config import ByteQuantity, addByteQuantityNode, readByteQuantity
from CedarBackup3.util import changeOwnership, executeCommand, resolveCommand
from CedarBackup3.xmlutil import addContainerNode, createInputDom, readFirstChild

########################################################################
# Module-wide constants and variables
########################################################################

logger = logging.getLogger("CedarBackup3.log.extend.split")

SPLIT_COMMAND = ["split"]
SPLIT_INDICATOR = "cback.split"


########################################################################
# SplitConfig class definition
########################################################################


@total_ordering
class SplitConfig:
    """
    Class representing split configuration.

    Split configuration is used for splitting staging directories.

    The following restrictions exist on data in this class:

       - The size limit must be a ByteQuantity
       - The split size must be a ByteQuantity

    """

    def __init__(self, sizeLimit=None, splitSize=None):
        """
        Constructor for the ``SplitCOnfig`` class.

        Args:
           sizeLimit: Size limit of the files, in bytes
           splitSize: Size that files exceeding the limit will be split into, in bytes

        Raises:
           ValueError: If one of the values is invalid
        """
        self._sizeLimit = None
        self._splitSize = None
        self.sizeLimit = sizeLimit
        self.splitSize = splitSize

    def __repr__(self):
        """
        Official string representation for class instance.
        """
        return "SplitConfig(%s, %s)" % (self.sizeLimit, self.splitSize)

    def __str__(self):
        """
        Informal string representation for class instance.
        """
        return self.__repr__()

    def __eq__(self, other):
        """Equals operator, iplemented in terms of original Python 2 compare operator."""
        return self.__cmp__(other) == 0

    def __lt__(self, other):
        """Less-than operator, iplemented in terms of original Python 2 compare operator."""
        return self.__cmp__(other) < 0

    def __gt__(self, other):
        """Greater-than operator, iplemented in terms of original Python 2 compare operator."""
        return self.__cmp__(other) > 0

    def __cmp__(self, other):
        """
        Original Python 2 comparison operator.
        Lists within this class are "unordered" for equality comparisons.
        Args:
           other: Other object to compare to
        Returns:
            -1/0/1 depending on whether self is ``<``, ``=`` or ``>`` other
        """
        if other is None:
            return 1
        if self.sizeLimit != other.sizeLimit:
            if (self.sizeLimit or ByteQuantity()) < (other.sizeLimit or ByteQuantity()):
                return -1
            else:
                return 1
        if self.splitSize != other.splitSize:
            if (self.splitSize or ByteQuantity()) < (other.splitSize or ByteQuantity()):
                return -1
            else:
                return 1
        return 0

    def _setSizeLimit(self, value):
        """
        Property target used to set the size limit.
        If not ``None``, the value must be a ``ByteQuantity`` object.
        Raises:
           ValueError: If the value is not a ``ByteQuantity``
        """
        if value is None:
            self._sizeLimit = None
        else:
            if not isinstance(value, ByteQuantity):
                raise ValueError("Value must be a ``ByteQuantity`` object.")
            self._sizeLimit = value

    def _getSizeLimit(self):
        """
        Property target used to get the size limit.
        """
        return self._sizeLimit

    def _setSplitSize(self, value):
        """
        Property target used to set the split size.
        If not ``None``, the value must be a ``ByteQuantity`` object.
        Raises:
           ValueError: If the value is not a ``ByteQuantity``
        """
        if value is None:
            self._splitSize = None
        else:
            if not isinstance(value, ByteQuantity):
                raise ValueError("Value must be a ``ByteQuantity`` object.")
            self._splitSize = value

    def _getSplitSize(self):
        """
        Property target used to get the split size.
        """
        return self._splitSize

    sizeLimit = property(_getSizeLimit, _setSizeLimit, None, doc="Size limit, as a ByteQuantity")
    splitSize = property(_getSplitSize, _setSplitSize, None, doc="Split size, as a ByteQuantity")


########################################################################
# LocalConfig class definition
########################################################################


@total_ordering
class LocalConfig:
    """
    Class representing this extension's configuration document.

    This is not a general-purpose configuration object like the main Cedar
    Backup configuration object.  Instead, it just knows how to parse and emit
    split-specific configuration values.  Third parties who need to read and
    write configuration related to this extension should access it through the
    constructor, ``validate`` and ``addConfig`` methods.

    *Note:* Lists within this class are "unordered" for equality comparisons.

    """

    def __init__(self, xmlData=None, xmlPath=None, validate=True):
        """
        Initializes a configuration object.

        If you initialize the object without passing either ``xmlData`` or
        ``xmlPath`` then configuration will be empty and will be invalid until it
        is filled in properly.

        No reference to the original XML data or original path is saved off by
        this class.  Once the data has been parsed (successfully or not) this
        original information is discarded.

        Unless the ``validate`` argument is ``False``, the :any:`LocalConfig.validate`
        method will be called (with its default arguments) against configuration
        after successfully parsing any passed-in XML.  Keep in mind that even if
        ``validate`` is ``False``, it might not be possible to parse the passed-in
        XML document if lower-level validations fail.

        *Note:* It is strongly suggested that the ``validate`` option always be set
        to ``True`` (the default) unless there is a specific need to read in
        invalid configuration from disk.

        Args:
           xmlData (String data): XML data representing configuration
           xmlPath (Absolute path to a file on disk): Path to an XML file on disk
           validate (Boolean true/false): Validate the document after parsing it
        Raises:
           ValueError: If both ``xmlData`` and ``xmlPath`` are passed-in
           ValueError: If the XML data in ``xmlData`` or ``xmlPath`` cannot be parsed
           ValueError: If the parsed configuration document is not valid
        """
        self._split = None
        self.split = None
        if xmlData is not None and xmlPath is not None:
            raise ValueError("Use either xmlData or xmlPath, but not both.")
        if xmlData is not None:
            self._parseXmlData(xmlData)
            if validate:
                self.validate()
        elif xmlPath is not None:
            with open(xmlPath) as f:
                xmlData = f.read()
            self._parseXmlData(xmlData)
            if validate:
                self.validate()

    def __repr__(self):
        """
        Official string representation for class instance.
        """
        return "LocalConfig(%s)" % (self.split)

    def __str__(self):
        """
        Informal string representation for class instance.
        """
        return self.__repr__()

    def __eq__(self, other):
        """Equals operator, iplemented in terms of original Python 2 compare operator."""
        return self.__cmp__(other) == 0

    def __lt__(self, other):
        """Less-than operator, iplemented in terms of original Python 2 compare operator."""
        return self.__cmp__(other) < 0

    def __gt__(self, other):
        """Greater-than operator, iplemented in terms of original Python 2 compare operator."""
        return self.__cmp__(other) > 0

    def __cmp__(self, other):
        """
        Original Python 2 comparison operator.
        Lists within this class are "unordered" for equality comparisons.
        Args:
           other: Other object to compare to
        Returns:
            -1/0/1 depending on whether self is ``<``, ``=`` or ``>`` other
        """
        if other is None:
            return 1
        if self.split != other.split:
            if self.split < other.split:
                return -1
            else:
                return 1
        return 0

    def _setSplit(self, value):
        """
        Property target used to set the split configuration value.
        If not ``None``, the value must be a ``SplitConfig`` object.
        Raises:
           ValueError: If the value is not a ``SplitConfig``
        """
        if value is None:
            self._split = None
        else:
            if not isinstance(value, SplitConfig):
                raise ValueError("Value must be a ``SplitConfig`` object.")
            self._split = value

    def _getSplit(self):
        """
        Property target used to get the split configuration value.
        """
        return self._split

    split = property(_getSplit, _setSplit, None, "Split configuration in terms of a ``SplitConfig`` object.")

    def validate(self):
        """
        Validates configuration represented by the object.

        Split configuration must be filled in.  Within that, both the size limit
        and split size must be filled in.

        Raises:
           ValueError: If one of the validations fails
        """
        if self.split is None:
            raise ValueError("Split section is required.")
        if self.split.sizeLimit is None:
            raise ValueError("Size limit must be set.")
        if self.split.splitSize is None:
            raise ValueError("Split size must be set.")

    def addConfig(self, xmlDom, parentNode):
        """
        Adds a <split> configuration section as the next child of a parent.

        Third parties should use this function to write configuration related to
        this extension.

        We add the following fields to the document::

           sizeLimit      //cb_config/split/size_limit
           splitSize      //cb_config/split/split_size

        Args:
           xmlDom: DOM tree as from ``impl.createDocument()``
           parentNode: Parent that the section should be appended to
        """
        if self.split is not None:
            sectionNode = addContainerNode(xmlDom, parentNode, "split")
            addByteQuantityNode(xmlDom, sectionNode, "size_limit", self.split.sizeLimit)
            addByteQuantityNode(xmlDom, sectionNode, "split_size", self.split.splitSize)

    def _parseXmlData(self, xmlData):
        """
        Internal method to parse an XML string into the object.

        This method parses the XML document into a DOM tree (``xmlDom``) and then
        calls a static method to parse the split configuration section.

        Args:
           xmlData (String data): XML data to be parsed
        Raises:
           ValueError: If the XML cannot be successfully parsed
        """
        (_, parentNode) = createInputDom(xmlData)
        self._split = LocalConfig._parseSplit(parentNode)

    @staticmethod
    def _parseSplit(parent):
        """
        Parses an split configuration section.

        We read the following individual fields::

           sizeLimit      //cb_config/split/size_limit
           splitSize      //cb_config/split/split_size

        Args:
           parent: Parent node to search beneath

        Returns:
            ``EncryptConfig`` object or ``None`` if the section does not exist
        Raises:
           ValueError: If some filled-in value is invalid
        """
        split = None
        section = readFirstChild(parent, "split")
        if section is not None:
            split = SplitConfig()
            split.sizeLimit = readByteQuantity(section, "size_limit")
            split.splitSize = readByteQuantity(section, "split_size")
        return split


########################################################################
# Public functions
########################################################################

###########################
# executeAction() function
###########################


def executeAction(configPath, options, config):  # noqa: ARG001
    """
    Executes the split backup action.

    Args:
       configPath (String representing a path on disk): Path to configuration file on disk
       options (Options object): Program command-line options
       config (Config object): Program configuration
    Raises:
       ValueError: Under many generic error conditions
       IOError: If there are I/O problems reading or writing files
    """
    logger.debug("Executing split extended action.")
    if config.options is None or config.stage is None:
        raise ValueError("Cedar Backup configuration is not properly filled in.")
    local = LocalConfig(xmlPath=configPath)
    dailyDirs = findDailyDirs(config.stage.targetDir, SPLIT_INDICATOR)
    for dailyDir in dailyDirs:
        _splitDailyDir(
            dailyDir, local.split.sizeLimit, local.split.splitSize, config.options.backupUser, config.options.backupGroup
        )
        writeIndicatorFile(dailyDir, SPLIT_INDICATOR, config.options.backupUser, config.options.backupGroup)
    logger.info("Executed the split extended action successfully.")


##############################
# _splitDailyDir() function
##############################


def _splitDailyDir(dailyDir, sizeLimit, splitSize, backupUser, backupGroup):
    """
    Splits large files in a daily staging directory.

    Files that match INDICATOR_PATTERNS (i.e. ``"cback.store"``,
    ``"cback.stage"``, etc.) are assumed to be indicator files and are ignored.
    All other files are split.

    Args:
       dailyDir: Daily directory to encrypt
       sizeLimit: Size limit, in bytes
       splitSize: Split size, in bytes
       backupUser: User that target files should be owned by
       backupGroup: Group that target files should be owned by

    Raises:
       ValueError: If the encrypt mode is not supported
       ValueError: If the daily staging directory does not exist
    """
    logger.debug("Begin splitting contents of [%s].", dailyDir)
    fileList = getBackupFiles(dailyDir)  # ignores indicator files
    for path in fileList:
        size = float(os.stat(path).st_size)
        if size > sizeLimit:
            _splitFile(path, splitSize, backupUser, backupGroup, removeSource=True)
    logger.debug("Completed splitting contents of [%s].", dailyDir)


########################
# _splitFile() function
########################


def _splitFile(sourcePath, splitSize, backupUser, backupGroup, removeSource=False):
    """
    Splits the source file into chunks of the indicated size.

    The split files will be owned by the indicated backup user and group.  If
    ``removeSource`` is ``True``, then the source file will be removed after it is
    successfully split.

    Args:
       sourcePath: Absolute path of the source file to split
       splitSize: Encryption mode (only "gpg" is allowed)
       backupUser: User that target files should be owned by
       backupGroup: Group that target files should be owned by
       removeSource: Indicates whether to remove the source file

    Raises:
       IOError: If there is a problem accessing, splitting or removing the source file
    """
    cwd = os.getcwd()
    try:
        if not os.path.exists(sourcePath):
            raise ValueError("Source path [%s] does not exist." % sourcePath)
        dirname = os.path.dirname(sourcePath)
        filename = os.path.basename(sourcePath)
        prefix = "%s_" % filename
        bytes = int(splitSize.bytes)  # noqa: A001
        os.chdir(dirname)  # need to operate from directory that we want files written to
        command = resolveCommand(SPLIT_COMMAND)
        args = ["--verbose", "--numeric-suffixes", "--suffix-length=5", "--bytes=%d" % bytes, filename, prefix]
        (result, output) = executeCommand(command, args, returnOutput=True, ignoreStderr=False)
        if result != 0:
            raise OSError("Error [%d] calling split for [%s]." % (result, sourcePath))
        pattern = re.compile(r"(creating file [`'])(%s)(.*)(')" % prefix)
        match = pattern.search(output[-1:][0])
        if match is None:
            raise OSError("Unable to parse output from split command.")
        value = int(match.group(3).strip())
        for index in range(value):
            path = "%s%05d" % (prefix, index)
            if not os.path.exists(path):
                raise OSError("After call to split, expected file [%s] does not exist." % path)
            changeOwnership(path, backupUser, backupGroup)
        if removeSource:
            if os.path.exists(sourcePath):
                try:
                    os.remove(sourcePath)
                    logger.debug("Completed removing old file [%s].", sourcePath)
                except Exception:
                    raise OSError("Failed to remove file [%s] after splitting it." % (sourcePath))
    finally:
        os.chdir(cwd)

1	# vim: set ft=python ts=4 sw=4 expandtab:
2	# # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # #
3	#
4	# C E D A R
5	# S O L U T I O N S "Software done right."
6	# S O F T W A R E
7	#
8	# # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # #
9	#
10	# Copyright (c) 2007,2010,2013,2015 Kenneth J. Pronovici.
11	# All rights reserved.
12	#
13	# This program is free software; you can redistribute it and/or
14	# modify it under the terms of the GNU General Public License,
15	# Version 2, as published by the Free Software Foundation.
16	#
17	# This program is distributed in the hope that it will be useful,
18	# but WITHOUT ANY WARRANTY; without even the implied warranty of
19	# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
20	#
21	# Copies of the GNU General Public License are available from
22	# the Free Software Foundation website, http://www.gnu.org/.
23	#
24	# # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # #
25	#
26	# Author : Kenneth J. Pronovici <pronovic@ieee.org>
27	# Language : Python 3
28	# Project : Official Cedar Backup Extensions
29	# Purpose : Provides an extension to split up large files in staging directories.
30	#
31	# # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # #
32
33	########################################################################
34	# Module documentation
35	########################################################################
36
37	"""
38	Provides an extension to split up large files in staging directories.
39
40	When this extension is executed, it will look through the configured Cedar
41	Backup staging directory for files exceeding a specified size limit, and split
42	them down into smaller files using the 'split' utility. Any directory which
43	has already been split (as indicated by the ``cback.split`` file) will be
44	ignored.
45
46	This extension requires a new configuration section <split> and is intended
47	to be run immediately after the standard stage action or immediately before the
48	standard store action. Aside from its own configuration, it requires the
49	options and staging configuration sections in the standard Cedar Backup
50	configuration file.
51
52	:author: Kenneth J. Pronovici <pronovic@ieee.org>
53	"""
54
55	########################################################################
56	# Imported modules
57	########################################################################
58
59	import logging	4✔
60	import os	4✔
61	import re	4✔
62	from functools import total_ordering	4✔
63
64	from CedarBackup3.actions.util import findDailyDirs, getBackupFiles, writeIndicatorFile	4✔
65	from CedarBackup3.config import ByteQuantity, addByteQuantityNode, readByteQuantity	4✔
66	from CedarBackup3.util import changeOwnership, executeCommand, resolveCommand	4✔
67	from CedarBackup3.xmlutil import addContainerNode, createInputDom, readFirstChild	4✔
68
69	########################################################################
70	# Module-wide constants and variables
71	########################################################################
72
73	logger = logging.getLogger("CedarBackup3.log.extend.split")	4✔
74
75	SPLIT_COMMAND = ["split"]	4✔
76	SPLIT_INDICATOR = "cback.split"	4✔
77
78
79	########################################################################
80	# SplitConfig class definition
81	########################################################################
82
83
84	@total_ordering	4✔
85	class SplitConfig:	4✔
86	"""
87	Class representing split configuration.
88
89	Split configuration is used for splitting staging directories.
90
91	The following restrictions exist on data in this class:
92
93	- The size limit must be a ByteQuantity
94	- The split size must be a ByteQuantity
95
96	"""
97
98	def __init__(self, sizeLimit=None, splitSize=None):	4✔
99	"""
100	Constructor for the ``SplitCOnfig`` class.
101
102	Args:
103	sizeLimit: Size limit of the files, in bytes
104	splitSize: Size that files exceeding the limit will be split into, in bytes
105
106	Raises:
107	ValueError: If one of the values is invalid
108	"""
109	self._sizeLimit = None	4✔
110	self._splitSize = None	4✔
111	self.sizeLimit = sizeLimit	4✔
112	self.splitSize = splitSize	4✔
113
114	def __repr__(self):	4✔
115	"""
116	Official string representation for class instance.
117	"""
118	return "SplitConfig(%s, %s)" % (self.sizeLimit, self.splitSize)	4✔
119
120	def __str__(self):	4✔
121	"""
122	Informal string representation for class instance.
123	"""
124	return self.__repr__()	4✔
125
126	def __eq__(self, other):	4✔
127	"""Equals operator, iplemented in terms of original Python 2 compare operator."""
128	return self.__cmp__(other) == 0	4✔
129
130	def __lt__(self, other):	4✔
131	"""Less-than operator, iplemented in terms of original Python 2 compare operator."""
132	return self.__cmp__(other) < 0	4✔
133
134	def __gt__(self, other):	4✔
135	"""Greater-than operator, iplemented in terms of original Python 2 compare operator."""
136	return self.__cmp__(other) > 0	4✔
137
138	def __cmp__(self, other):	4✔
139	"""
140	Original Python 2 comparison operator.
141	Lists within this class are "unordered" for equality comparisons.
142	Args:
143	other: Other object to compare to
144	Returns:
145	-1/0/1 depending on whether self is ``<``, ``=`` or ``>`` other
146	"""
147	if other is None:	4✔
148	return 1	4✔
149	if self.sizeLimit != other.sizeLimit:	4✔
150	if (self.sizeLimit or ByteQuantity()) < (other.sizeLimit or ByteQuantity()):	4✔
151	return -1	4✔
152	else:
153	return 1	×
154	if self.splitSize != other.splitSize:	4✔
155	if (self.splitSize or ByteQuantity()) < (other.splitSize or ByteQuantity()):	4✔
156	return -1	4✔
157	else:
158	return 1	×
159	return 0	4✔
160
161	def _setSizeLimit(self, value):	4✔
162	"""
163	Property target used to set the size limit.
164	If not ``None``, the value must be a ``ByteQuantity`` object.
165	Raises:
166	ValueError: If the value is not a ``ByteQuantity``
167	"""
168	if value is None:	4✔
169	self._sizeLimit = None	4✔
170	else:
171	if not isinstance(value, ByteQuantity):	4✔
172	raise ValueError("Value must be a ``ByteQuantity`` object.")	4✔
173	self._sizeLimit = value	4✔
174
175	def _getSizeLimit(self):	4✔
176	"""
177	Property target used to get the size limit.
178	"""
179	return self._sizeLimit	4✔
180
181	def _setSplitSize(self, value):	4✔
182	"""
183	Property target used to set the split size.
184	If not ``None``, the value must be a ``ByteQuantity`` object.
185	Raises:
186	ValueError: If the value is not a ``ByteQuantity``
187	"""
188	if value is None:	4✔
189	self._splitSize = None	4✔
190	else:
191	if not isinstance(value, ByteQuantity):	4✔
192	raise ValueError("Value must be a ``ByteQuantity`` object.")	4✔
193	self._splitSize = value	4✔
194
195	def _getSplitSize(self):	4✔
196	"""
197	Property target used to get the split size.
198	"""
199	return self._splitSize	4✔
200
201	sizeLimit = property(_getSizeLimit, _setSizeLimit, None, doc="Size limit, as a ByteQuantity")	4✔
202	splitSize = property(_getSplitSize, _setSplitSize, None, doc="Split size, as a ByteQuantity")	4✔
203
204
205	########################################################################
206	# LocalConfig class definition
207	########################################################################
208
209
210	@total_ordering	4✔
211	class LocalConfig:	4✔
212	"""
213	Class representing this extension's configuration document.
214
215	This is not a general-purpose configuration object like the main Cedar
216	Backup configuration object. Instead, it just knows how to parse and emit
217	split-specific configuration values. Third parties who need to read and
218	write configuration related to this extension should access it through the
219	constructor, ``validate`` and ``addConfig`` methods.
220
221	Note: Lists within this class are "unordered" for equality comparisons.
222
223	"""
224
225	def __init__(self, xmlData=None, xmlPath=None, validate=True):	4✔
226	"""
227	Initializes a configuration object.
228
229	If you initialize the object without passing either ``xmlData`` or
230	``xmlPath`` then configuration will be empty and will be invalid until it
231	is filled in properly.
232
233	No reference to the original XML data or original path is saved off by
234	this class. Once the data has been parsed (successfully or not) this
235	original information is discarded.
236
237	Unless the ``validate`` argument is ``False``, the :any:`LocalConfig.validate`
238	method will be called (with its default arguments) against configuration
239	after successfully parsing any passed-in XML. Keep in mind that even if
240	``validate`` is ``False``, it might not be possible to parse the passed-in
241	XML document if lower-level validations fail.
242
243	Note: It is strongly suggested that the ``validate`` option always be set
244	to ``True`` (the default) unless there is a specific need to read in
245	invalid configuration from disk.
246
247	Args:
248	xmlData (String data): XML data representing configuration
249	xmlPath (Absolute path to a file on disk): Path to an XML file on disk
250	validate (Boolean true/false): Validate the document after parsing it
251	Raises:
252	ValueError: If both ``xmlData`` and ``xmlPath`` are passed-in
253	ValueError: If the XML data in ``xmlData`` or ``xmlPath`` cannot be parsed
254	ValueError: If the parsed configuration document is not valid
255	"""
256	self._split = None	4✔
257	self.split = None	4✔
258	if xmlData is not None and xmlPath is not None:	4✔
259	raise ValueError("Use either xmlData or xmlPath, but not both.")	4✔
260	if xmlData is not None:	4✔
261	self._parseXmlData(xmlData)	4✔
262	if validate:	4✔
263	self.validate()	4✔
264	elif xmlPath is not None:	4✔
265	with open(xmlPath) as f:	4✔
266	xmlData = f.read()	4✔
267	self._parseXmlData(xmlData)	4✔
268	if validate:	4✔
269	self.validate()	4✔
270
271	def __repr__(self):	4✔
272	"""
273	Official string representation for class instance.
274	"""
275	return "LocalConfig(%s)" % (self.split)	4✔
276
277	def __str__(self):	4✔
278	"""
279	Informal string representation for class instance.
280	"""
281	return self.__repr__()	4✔
282
283	def __eq__(self, other):	4✔
284	"""Equals operator, iplemented in terms of original Python 2 compare operator."""
285	return self.__cmp__(other) == 0	4✔
286
287	def __lt__(self, other):	4✔
288	"""Less-than operator, iplemented in terms of original Python 2 compare operator."""
289	return self.__cmp__(other) < 0	4✔
290
291	def __gt__(self, other):	4✔
292	"""Greater-than operator, iplemented in terms of original Python 2 compare operator."""
293	return self.__cmp__(other) > 0	4✔
294
295	def __cmp__(self, other):	4✔
296	"""
297	Original Python 2 comparison operator.
298	Lists within this class are "unordered" for equality comparisons.
299	Args:
300	other: Other object to compare to
301	Returns:
302	-1/0/1 depending on whether self is ``<``, ``=`` or ``>`` other
303	"""
304	if other is None:	4✔
305	return 1	×
306	if self.split != other.split:	4✔
307	if self.split < other.split:	4✔
308	return -1	4✔
309	else:
310	return 1	×
311	return 0	4✔
312
313	def _setSplit(self, value):	4✔
314	"""
315	Property target used to set the split configuration value.
316	If not ``None``, the value must be a ``SplitConfig`` object.
317	Raises:
318	ValueError: If the value is not a ``SplitConfig``
319	"""
320	if value is None:	4✔
321	self._split = None	4✔
322	else:
323	if not isinstance(value, SplitConfig):	4✔
324	raise ValueError("Value must be a ``SplitConfig`` object.")	4✔
325	self._split = value	4✔
326
327	def _getSplit(self):	4✔
328	"""
329	Property target used to get the split configuration value.
330	"""
331	return self._split	4✔
332
333	split = property(_getSplit, _setSplit, None, "Split configuration in terms of a ``SplitConfig`` object.")	4✔
334
335	def validate(self):	4✔
336	"""
337	Validates configuration represented by the object.
338
339	Split configuration must be filled in. Within that, both the size limit
340	and split size must be filled in.
341
342	Raises:
343	ValueError: If one of the validations fails
344	"""
345	if self.split is None:	4✔
346	raise ValueError("Split section is required.")	4✔
347	if self.split.sizeLimit is None:	4✔
348	raise ValueError("Size limit must be set.")	4✔
349	if self.split.splitSize is None:	4✔
350	raise ValueError("Split size must be set.")	4✔
351
352	def addConfig(self, xmlDom, parentNode):	4✔
353	"""
354	Adds a <split> configuration section as the next child of a parent.
355
356	Third parties should use this function to write configuration related to
357	this extension.
358
359	We add the following fields to the document::
360
361	sizeLimit //cb_config/split/size_limit
362	splitSize //cb_config/split/split_size
363
364	Args:
365	xmlDom: DOM tree as from ``impl.createDocument()``
366	parentNode: Parent that the section should be appended to
367	"""
368	if self.split is not None:	4✔
369	sectionNode = addContainerNode(xmlDom, parentNode, "split")	4✔
370	addByteQuantityNode(xmlDom, sectionNode, "size_limit", self.split.sizeLimit)	4✔
371	addByteQuantityNode(xmlDom, sectionNode, "split_size", self.split.splitSize)	4✔
372
373	def _parseXmlData(self, xmlData):	4✔
374	"""
375	Internal method to parse an XML string into the object.
376
377	This method parses the XML document into a DOM tree (``xmlDom``) and then
378	calls a static method to parse the split configuration section.
379
380	Args:
381	xmlData (String data): XML data to be parsed
382	Raises:
383	ValueError: If the XML cannot be successfully parsed
384	"""
385	(_, parentNode) = createInputDom(xmlData)	4✔
386	self._split = LocalConfig._parseSplit(parentNode)	4✔
387
388	@staticmethod	4✔
389	def _parseSplit(parent):	4✔
390	"""
391	Parses an split configuration section.
392
393	We read the following individual fields::
394
395	sizeLimit //cb_config/split/size_limit
396	splitSize //cb_config/split/split_size
397
398	Args:
399	parent: Parent node to search beneath
400
401	Returns:
402	``EncryptConfig`` object or ``None`` if the section does not exist
403	Raises:
404	ValueError: If some filled-in value is invalid
405	"""
406	split = None	4✔
407	section = readFirstChild(parent, "split")	4✔
408	if section is not None:	4✔
409	split = SplitConfig()	4✔
410	split.sizeLimit = readByteQuantity(section, "size_limit")	4✔
411	split.splitSize = readByteQuantity(section, "split_size")	4✔
412	return split	4✔
413
414
415	########################################################################
416	# Public functions
417	########################################################################
418
419	###########################
420	# executeAction() function
421	###########################
422
423
424	def executeAction(configPath, options, config): # noqa: ARG001	4✔
425	"""
426	Executes the split backup action.
427
428	Args:
429	configPath (String representing a path on disk): Path to configuration file on disk
430	options (Options object): Program command-line options
431	config (Config object): Program configuration
432	Raises:
433	ValueError: Under many generic error conditions
434	IOError: If there are I/O problems reading or writing files
435	"""
436	logger.debug("Executing split extended action.")	×
437	if config.options is None or config.stage is None:	×
438	raise ValueError("Cedar Backup configuration is not properly filled in.")	×
439	local = LocalConfig(xmlPath=configPath)	×
440	dailyDirs = findDailyDirs(config.stage.targetDir, SPLIT_INDICATOR)	×
441	for dailyDir in dailyDirs:	×
442	_splitDailyDir(	×
443	dailyDir, local.split.sizeLimit, local.split.splitSize, config.options.backupUser, config.options.backupGroup
444	)
445	writeIndicatorFile(dailyDir, SPLIT_INDICATOR, config.options.backupUser, config.options.backupGroup)	×
446	logger.info("Executed the split extended action successfully.")	×
447
448
449	##############################
450	# _splitDailyDir() function
451	##############################
452
453
454	def _splitDailyDir(dailyDir, sizeLimit, splitSize, backupUser, backupGroup):	4✔
455	"""
456	Splits large files in a daily staging directory.
457
458	Files that match INDICATOR_PATTERNS (i.e. ``"cback.store"``,
459	``"cback.stage"``, etc.) are assumed to be indicator files and are ignored.
460	All other files are split.
461
462	Args:
463	dailyDir: Daily directory to encrypt
464	sizeLimit: Size limit, in bytes
465	splitSize: Split size, in bytes
466	backupUser: User that target files should be owned by
467	backupGroup: Group that target files should be owned by
468
469	Raises:
470	ValueError: If the encrypt mode is not supported
471	ValueError: If the daily staging directory does not exist
472	"""
473	logger.debug("Begin splitting contents of [%s].", dailyDir)	×
474	fileList = getBackupFiles(dailyDir) # ignores indicator files	×
475	for path in fileList:	×
476	size = float(os.stat(path).st_size)	×
477	if size > sizeLimit:	×
478	_splitFile(path, splitSize, backupUser, backupGroup, removeSource=True)	×
479	logger.debug("Completed splitting contents of [%s].", dailyDir)	×
480
481
482	########################
483	# _splitFile() function
484	########################
485
486
487	def _splitFile(sourcePath, splitSize, backupUser, backupGroup, removeSource=False):	4✔
488	"""
489	Splits the source file into chunks of the indicated size.
490
491	The split files will be owned by the indicated backup user and group. If
492	``removeSource`` is ``True``, then the source file will be removed after it is
493	successfully split.
494
495	Args:
496	sourcePath: Absolute path of the source file to split
497	splitSize: Encryption mode (only "gpg" is allowed)
498	backupUser: User that target files should be owned by
499	backupGroup: Group that target files should be owned by
500	removeSource: Indicates whether to remove the source file
501
502	Raises:
503	IOError: If there is a problem accessing, splitting or removing the source file
504	"""
505	cwd = os.getcwd()	×
506	try:	×
507	if not os.path.exists(sourcePath):	×
508	raise ValueError("Source path [%s] does not exist." % sourcePath)	×
509	dirname = os.path.dirname(sourcePath)	×
510	filename = os.path.basename(sourcePath)	×
511	prefix = "%s_" % filename	×
NEW 512	bytes = int(splitSize.bytes) # noqa: A001	×
513	os.chdir(dirname) # need to operate from directory that we want files written to	×
514	command = resolveCommand(SPLIT_COMMAND)	×
515	args = ["--verbose", "--numeric-suffixes", "--suffix-length=5", "--bytes=%d" % bytes, filename, prefix]	×
516	(result, output) = executeCommand(command, args, returnOutput=True, ignoreStderr=False)	×
517	if result != 0:	×
NEW 518	raise OSError("Error [%d] calling split for [%s]." % (result, sourcePath))	×
519	pattern = re.compile(r"(creating file [`'])(%s)(.*)(')" % prefix)	×
520	match = pattern.search(output[-1:][0])	×
521	if match is None:	×
NEW 522	raise OSError("Unable to parse output from split command.")	×
523	value = int(match.group(3).strip())	×
NEW 524	for index in range(value):	×
525	path = "%s%05d" % (prefix, index)	×
526	if not os.path.exists(path):	×
NEW 527	raise OSError("After call to split, expected file [%s] does not exist." % path)	×
528	changeOwnership(path, backupUser, backupGroup)	×
529	if removeSource:	×
530	if os.path.exists(sourcePath):	×
531	try:	×
532	os.remove(sourcePath)	×
533	logger.debug("Completed removing old file [%s].", sourcePath)	×
NEW 534	except Exception:	×
NEW 535	raise OSError("Failed to remove file [%s] after splitting it." % (sourcePath))	×
536	finally:
537	os.chdir(cwd)	×

pronovic / cedar-backup3 / 17816721342

Source File Press 'n' to go to next uncovered line, 'b' for previous

Source File
Press 'n' to go to next uncovered line, 'b' for previous