Commit a8530cd9 authored by cvs2svn's avatar cvs2svn

This commit was manufactured by cvs2svn to create tag 'r0-10-2'.

git-svn-id: http://svn.savannah.nongnu.org/svn/rdiff-backup@266 2b77aa54-bcbc-44c9-a7ec-4f6cf2b41109
parent 426cc1c0
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
INSTALLATION:
Thank you for trying rdiff-backup. To install, run:
python setup.py install
The build process can be also be run separately:
python setup.py build
The default prefix is /usr, so files are put in /usr/bin,
/usr/share/man/, etc. An alternate prefix can be specified using the
--prefix=<prefix> option. For example:
python setup.py install --prefix=/usr/local
REQUIREMENTS:
Remember that you must have Python 2.2 or later and librsync 0.9.5.1
or later installed. For Python, see http://www.python.org. The
rdiff-backup homepage at http://rdiff-backup.stanford.edu/ should have
a recent version of librsync; otherwise see the librsync homepage at
http://sourceforge.net/projects/librsync/.
For remote operation, rdiff-backup should be in installed and in the
PATH on remote system(s) (see man page for more information).
TROUBLESHOOTING:
If you have everything installed properly, and it still doesn't work,
see the enclosed FAQ.html, the web page at
http://rdiff-backup.stanford.edu, and/or the mailing list.
Update FAQ on cstream
Write some better selection test cases to test new Iterate_fast func.
Work on killtest code - avoid returning a failure when a file is
simply skipped.
---------[ Medium term ]---------------------------------------
Interpret a list of filenames in globbing form
(--include-globbing-filelist maybe).
Add --dry-run option (target for v1.1.x)
Add # of increments option to --remove-older-than
Restore only changed files
Make argument shortcut for cstream
Make --calculate-averages work with directory_statistics file.
---------[ Long term ]---------------------------------------
Think about adding Gaudet's idea for keeping track of renamed files.
Look into different inode generation techniques (see treescan, Dean
Gaudet's other post).
#!/usr/bin/env python
import os, re, shutil, time, sys
SourceDir = "src"
DistDir = "dist"
# Various details about the files must also be specified by the rpm
# spec template.
spec_template = "dist/rdiff-backup.spec"
#redhat_spec_template = "dist/rdiff-backup.rh7x.spec"
def CopyMan(destination, version):
"""Create updated man page at the specified location"""
fp = open(destination, "w")
date = time.strftime("%B %Y", time.localtime(time.time()))
version = "Version "+version
firstline = ('.TH RDIFF-BACKUP 1 "%s" "%s" "User Manuals"\n' %
(date, version))
fp.write(firstline)
infp = open("rdiff-backup.1", "r")
infp.readline()
fp.write(infp.read())
fp.close()
infp.close()
def MakeFAQ():
"""Create FAQ.html and FAQ.wml files from FAQ-body.html"""
faqbody_fp = open("FAQ-body.html", "r")
faqbody_string = faqbody_fp.read()
faqbody_fp.close()
wml_fp = open("FAQ.wml", "w")
wml_fp.write(
"""#include 'template.wml' curpage=faq title="rdiff-backup: FAQ"
<divert body>
<p><h2>FAQ:</h2>
""")
wml_fp.write(faqbody_string)
wml_fp.write("\n</divert>\n")
wml_fp.close()
html_fp = open("FAQ.html", "w")
html_fp.write(
"""<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<title>rdiff-backup FAQ</title>
</head>
<body>
<h1>rdiff-backup FAQ</h1>
""")
html_fp.write(faqbody_string)
html_fp.write("\n</body></html>")
html_fp.close()
def VersionedCopy(source, dest):
"""Copy source to dest, substituting $version with version"""
fin = open(source, "rb")
inbuf = fin.read()
assert not fin.close()
outbuf = re.sub("\$version", Version, inbuf, 1)
if outbuf == inbuf: assert 0, "No $version string replaced"
assert not re.search("\$version", outbuf), \
"Two $version strings found in the same file %s" % (source,)
fout = open(dest, "wb")
fout.write(outbuf)
assert not fout.close()
def MakeTar():
"""Create rdiff-backup tar file"""
tardir = "rdiff-backup-%s" % Version
tarfile = "rdiff-backup-%s.tar.gz" % Version
try:
os.lstat(tardir)
os.system("rm -rf " + tardir)
except OSError: pass
os.mkdir(tardir)
for filename in ["CHANGELOG", "COPYING", "README", "FAQ.html",
SourceDir + "/cmodule.c",
SourceDir + "/_librsyncmodule.c",
DistDir + "/setup.py"]:
assert not os.system("cp %s %s" % (filename, tardir)), filename
os.mkdir(tardir+"/rdiff_backup")
for filename in ["connection.py", "destructive_stepping.py",
"FilenameMapping.py", "Hardlink.py",
"highlevel.py", "increment.py", "__init__.py",
"iterfile.py", "lazy.py", "librsync.py",
"log.py", "Main.py", "manage.py", "MiscStats.py",
"Rdiff.py", "restore.py", "rlist.py",
"robust.py", "rorpiter.py", "rpath.py", "Security.py",
"selection.py", "SetConnections.py", "static.py",
"statistics.py", "Time.py"]:
assert not os.system("cp %s/%s %s/rdiff_backup" %
(SourceDir, filename, tardir)), filename
VersionedCopy("%s/Globals.py" % (SourceDir,),
"%s/rdiff_backup/Globals.py" % (tardir,))
VersionedCopy("rdiff-backup", "%s/rdiff-backup" % (tardir,))
VersionedCopy(DistDir + "/setup.py", "%s/setup.py" % (tardir,))
os.chmod(os.path.join(tardir, "setup.py"), 0755)
os.chmod(os.path.join(tardir, "rdiff-backup"), 0644)
CopyMan(os.path.join(tardir, "rdiff-backup.1"), Version)
os.system("tar -cvzf %s %s" % (tarfile, tardir))
shutil.rmtree(tardir)
return tarfile
def MakeSpecFile():
"""Create spec file using spec template"""
specfile = "rdiff-backup-%s-1.spec" % Version
VersionedCopy(spec_template, specfile)
return specfile
def Main():
print "Making FAQ"
MakeFAQ()
print "Processing version " + Version
tarfile = MakeTar()
print "Made tar file " + tarfile
specfile = MakeSpecFile()
print "Made specfile ", specfile
if __name__ == "__main__" and not globals().has_key('__no_execute__'):
if len(sys.argv) != 2:
print "Syntax: makedist [version_number]"
sys.exit(1)
Version = sys.argv[1]
Main()
#!/usr/bin/env python
import os, sys, re
SourceDir = "src"
if len(sys.argv) == 2:
version = sys.argv[1]
specfile = "rdiff-backup-%s-1.spec" % version
print "Using specfile %s" % specfile
else:
print "Syntax: %s version_number" % sys.argv[0]
sys.exit(1)
base = ".".join(specfile.split(".")[:-1])
srcrpm = base+".src.rpm"
i386rpm = base+".i386.rpm"
source_rpm = base+".src.rpm"
tarfile = "-".join(base.split("-")[:-1]) + ".tar.gz"
# These assume the rpm root directory $HOME/rpm. The
# nonstandard location allows for building by non-root user.
assert not os.system("cp %s $HOME/rpm/SOURCES" % (tarfile,))
assert not os.system("rpmbuild -ba --sign -vv --target i386 " + specfile)
assert not os.system("cp $HOME/rpm/RPMS/i386/%s ." % i386rpm)
assert not os.system("cp $HOME/rpm/SRPMS/%s ." % source_rpm)
# Old root RPM instructions
#assert not os.system("install -o root -g root -m 644 %s "
# "/usr/src/redhat/SOURCES" % (tarfile,))
#assert not os.system("install -o ben -g ben -m 644 "
# "/usr/src/redhat/RPMS/i386/%s ." % i386rpm)
#assert not os.system("install -o ben -g ben -m 644 "
# "/usr/src/redhat/SRPMS/%s ." % source_rpm)
#!/usr/bin/env python
import sys, os
def RunCommand(cmd):
print cmd
os.system(cmd)
webprefix = "/home/ben/misc/html/mirror/rdiff-backup/"
if not sys.argv[1:]:
print 'Call with version number, as in "./makeweb 0.3.1"'
print "to move new rpms and tarballs. Now just remaking FAQ and man page."
print
else:
version = sys.argv[1]
RunCommand("cp *%s* %s" % (version, webprefix))
RunCommand("rman -f html -r '' rdiff-backup.1 > %srdiff-backup.1.html"
% webprefix)
RunCommand("cp FAQ.wml CHANGELOG %s" % webprefix)
os.chdir(webprefix)
print "cd ", webprefix
if sys.argv[1:]:
RunCommand("rm latest latest.rpm latest.tar.gz")
RunCommand("ln -s *%s*rpm latest.rpm" % (version,))
RunCommand("ln -s *%s*tar.gz latest.tar.gz" % (version,))
RunCommand("./Make")
--- rdiff-backup.old Sat Apr 6 10:05:18 2002
+++ rdiff-backup Sat Apr 6 10:05:25 2002
@@ -1,4 +1,4 @@
-#!/usr/bin/env python
+#!/usr/bin/env python2
#
# rdiff-backup -- Mirror files while keeping incremental changes
# Version 0.7.1 released March 25, 2002
Summary: Convenient and transparent local/remote incremental mirror/backup
Name: rdiff-backup
Release: 1
URL: http://www.stanford.edu/~bescoto/rdiff-backup/
Source: %{name}-%{version}.tar.gz
Copyright: GPL
Group: Applications/Archiving
BuildRoot: %{_tmppath}/%{name}-root
requires: librsync, python2 >= 2.2
Patch: rdiff-backup-rh7x.patch
%description
rdiff-backup is a script, written in Python, that backs up one
directory to another and is intended to be run periodically (nightly
from cron for instance). The target directory ends up a copy of the
source directory, but extra reverse diffs are stored in the target
directory, so you can still recover files lost some time ago. The idea
is to combine the best features of a mirror and an incremental
backup. rdiff-backup can also operate in a bandwidth efficient manner
over a pipe, like rsync. Thus you can use rdiff-backup and ssh to
securely back a hard drive up to a remote location, and only the
differences from the previous backup will be transmitted.
%prep
%setup
%patch
%build
%install
rm -rf $RPM_BUILD_ROOT
mkdir -p $RPM_BUILD_ROOT/usr/bin
mkdir -p $RPM_BUILD_ROOT/usr/share/man/man1
install -m 755 rdiff-backup $RPM_BUILD_ROOT/usr/bin/rdiff-backup
install -m 644 rdiff-backup.1 $RPM_BUILD_ROOT/usr/share/man/man1/rdiff-backup.1
%clean
%files
%defattr(-,root,root)
/usr/bin/rdiff-backup
/usr/share/man/man1/rdiff-backup.1.gz
%doc CHANGELOG COPYING README FAQ.html
%changelog
* Sat Apr 6 2002 Ben Escoto <bescoto@stanford.edu>
- Made new version for Redhat 7.x series
* Sun Nov 4 2001 Ben Escoto <bescoto@stanford.edu>
- Initial RPM
Version: $version
Summary: Convenient and transparent local/remote incremental mirror/backup
Name: rdiff-backup
Release: 1
URL: http://www.stanford.edu/~bescoto/rdiff-backup/
Source: %{name}-%{version}.tar.gz
Copyright: GPL
Group: Applications/Archiving
BuildRoot: %{_tmppath}/%{name}-root
requires: librsync >= 0.9.5.1, python2 >= 2.2
BuildPrereq: python2-devel >= 2.2, librsync-devel >= 0.9.5.1
%description
rdiff-backup is a script, written in Python, that backs up one
directory to another and is intended to be run periodically (nightly
from cron for instance). The target directory ends up a copy of the
source directory, but extra reverse diffs are stored in the target
directory, so you can still recover files lost some time ago. The idea
is to combine the best features of a mirror and an incremental
backup. rdiff-backup can also operate in a bandwidth efficient manner
over a pipe, like rsync. Thus you can use rdiff-backup and ssh to
securely back a hard drive up to a remote location, and only the
differences from the previous backup will be transmitted.
%prep
%setup
%build
python2 setup.py build
%install
python2 setup.py install --prefix=$RPM_BUILD_ROOT/usr
%clean
%files
%defattr(-,root,root)
/usr/bin/rdiff-backup
/usr/share/doc/rdiff-backup-%{version}
/usr/share/man/man1
/usr/lib
%changelog
* Sun Nov 4 2001 Ben Escoto <bescoto@stanford.edu>
- Initial RPM
#!/usr/bin/env python
import sys, os, getopt
from distutils.core import setup, Extension
version_string = "$version"
if sys.version_info[:2] < (2,2):
print "Sorry, rdiff-backup requires version 2.2 or later of python"
sys.exit(1)
setup(name="rdiff-backup",
version=version_string,
description="Local/remote mirroring+incremental backup",
author="Ben Escoto",
author_email="bescoto@stanford.edu",
url="http://rdiff-backup.stanford.edu",
packages = ['rdiff_backup'],
ext_modules = [Extension("rdiff_backup.C", ["cmodule.c"]),
Extension("rdiff_backup._librsync",
["_librsyncmodule.c"],
libraries=["rsync"])],
scripts = ['rdiff-backup'],
data_files = [('share/man/man1', ['rdiff-backup.1']),
('share/doc/rdiff-backup-%s' % version_string,
['CHANGELOG', 'COPYING', 'README', 'FAQ.html'])])
#!/usr/bin/env python
#
# Compresses old rdiff-backup increments. See
# http://www.stanford.edu/~bescoto/rdiff-backup for information on
# rdiff-backup.
from __future__ import nested_scopes, generators
import os, sys, getopt
rdiff_backup_location = "/usr/bin/rdiff-backup"
no_compression_regexp_string = None
__no_execute__ = 1
def print_help():
"""Print usage, exit"""
print """
Usage: compress-rdiff-backup-increments [options] mirror_directory
This script will compress the old rdiff-backup increments under
mirror_directory, in a format compatible with rdiff-backup version
0.7.1 and later. So for instance if you were using an old version
of rdiff-backup like this:
rdiff-backup /foo /backup
and now you want to take advantage of v0.7.1's space saving
compression, you can run:
compress-rdiff-backup-increments /backup
Options:
--rdiff-backup-location location
This script reads your rdiff-backup executable. The default
is "/usr/bin/rdiff-backup", so if your rdiff-backup is in a
different location, you must use this switch.
--no-compression-regexp regexp
Any increments whose base name match this regular expression
won't be compressed. This is generally used to avoid
compressing already compressed files. See the rdiff-backup
man page for the default.
"""
sys.exit(1)
def parse_args(arglist):
"""Check and evaluate command line arguments, return dirname"""
global rdiff_backup_location
global no_compression_regexp_string
try: optlist, args = getopt.getopt(arglist, "v:",
["rdiff-backup-location=",
"no-compression-regexp="])
except getopt.error: print_help()
for opt, arg in optlist:
if opt == "--no-compression-regexp":
no_compression_regexp_string = arg
elif opt == "--rdiff-backup-location": rdiff_backup_location = arg
else:
print "Bad option: ", opt
print_help()
if len(args) != 1:
print "Wrong number of arguments"
print_help()
return args[0]
def exec_rdiff_backup():
"""Execs rdiff-backup"""
try: execfile(rdiff_backup_location, globals())
except IOError:
print "Unable to read", rdiff_backup_location
print "You may need to use the --rdiff-backup-location argument"
sys.exit(1)
if not map(int, Globals.version.split(".")) >= [0, 7, 1]:
print "This script requires rdiff-backup version 0.7.1 or later,",
print "found version", Globals.version
sys.exit(1)
def gzip_file(rp):
"""gzip rp, adding .gz to path and deleting original"""
newrp = RPath(rp.conn, rp.base, rp.index[:-1] + (rp.index[-1]+".gz",))
if newrp.lstat():
print "Warning: %s already exists, skipping" % newrp.path
return
print "gzipping ", rp.path
newrp.write_from_fileobj(rp.open("rb"), compress = 1)
RPath.copy_attribs(rp, newrp)
rp.delete()
def Main():
dirname = parse_args(sys.argv[1:])
exec_rdiff_backup()
if no_compression_regexp_string is not None:
no_compression_regexp = re.compile(no_compression_regexp_string, re.I)
else: no_compression_regexp = \
re.compile(Globals.no_compression_regexp_string, re.I)
Globals.change_source_perms = 1
Globals.change_ownership = (os.getuid() == 0)
# Check to make sure rbdir exists
root_rp = RPath(Globals.local_connection, dirname)
rbdir = root_rp.append("rdiff-backup-data")
if not rbdir.lstat():
print "Cannot find %s, exiting" % rbdir.path
sys.exit(1)
for dsrp in DestructiveStepping.Iterate_with_Finalizer(rbdir, 1):
if (dsrp.isincfile() and dsrp.isreg() and
not dsrp.isinccompressed() and
(dsrp.getinctype() == "diff" or dsrp.getinctype() == "snapshot")
and dsrp.getsize() != 0 and
not no_compression_regexp.match(dsrp.getincbase_str())):
gzip_file(dsrp)
Main()
#!/usr/bin/env python
from __future__ import generators
import sys, os, stat
def usage():
print "Usage: find2dirs dir1 dir2"
print
print "Given the name of two directories, list all the files in both, one"
print "per line, but don't repeat a file even if it is in both directories"
sys.exit(1)
def getlist(base, ext = ""):
"""Return iterator yielding filenames from directory"""
if ext: yield ext
else: yield "."
fullname = os.path.join(base, ext)
if stat.S_ISDIR(stat.S_IFMT(os.lstat(fullname)[stat.ST_MODE])):
for subfile in os.listdir(fullname):
for fn in getlist(base, os.path.join(ext, subfile)): yield fn
def main(dir1, dir2):
d = {}
for fn in getlist(dir1): d[fn] = 1
for fn in getlist(dir2): d[fn] = 1
for fn in d.keys(): print fn
if not len(sys.argv) == 3: usage()
else: main(sys.argv[1], sys.argv[2])
#!/usr/bin/env python
"""init_smallfiles.py
This program makes a number of files of the given size in the
specified directory.
"""
import os, stat, sys, math
if len(sys.argv) > 5 or len(sys.argv) < 4:
print "Usage: init_files [directory name] [file size] [file count] [base]"
print
print "Creates file_count files in directory_name of size file_size."
print "The created directory has a tree type structure where each level"
print "has at most base files or directories in it. Default is 50."
sys.exit(1)
dirname = sys.argv[1]
filesize = int(sys.argv[2])
filecount = int(sys.argv[3])
block_size = 16384
block = "." * block_size
block_change = "." * (filesize % block_size)
if len(sys.argv) == 4: base = 50
else: base = int(sys.argv[4])
def make_file(path):
"""Make the file at path"""
fp = open(path, "w")
for i in xrange(int(math.floor(filesize/block_size))): fp.write(block)
fp.write(block_change)
fp.close()
def find_sublevels(count):
"""Return number of sublevels required for count files"""
return int(math.ceil(math.log(count)/math.log(base)))
def make_dir(dir, count):
"""Make count files in the directory, making subdirectories if necessary"""
print "Making directory %s with %d files" % (dir, count)
os.mkdir(dir)
level = find_sublevels(count)
assert count <= pow(base, level)
if level == 1:
for i in range(count): make_file(os.path.join(dir, "file%d" %i))
else:
files_per_subdir = pow(base, level-1)
full_dirs = int(count/files_per_subdir)
assert full_dirs <= base
for i in range(full_dirs):
make_dir(os.path.join(dir, "subdir%d" % i), files_per_subdir)
change = count - full_dirs*files_per_subdir
assert change >= 0
if change > 0:
make_dir(os.path.join(dir, "subdir%d" % full_dirs), change)
def start(dir):
try: os.stat(dir)
except os.error: pass
else:
print "Directory %s already exists, exiting." % dir
sys.exit(1)
make_dir(dirname, filecount)
start(dirname)
#!/usr/bin/env python
"""Use librsync to transform everything in one dir to another"""
import sys, os, librsync
dir1, dir2 = sys.argv[1:3]
for i in xrange(1000):
dir1fn = "%s/%s" % (dir1, i)
dir2fn = "%s/%s" % (dir2, i)
# Write signature file
f1 = open(dir1fn, "rb")
sigfile = open("sig", "wb")
librsync.filesig(f1, sigfile, 2048)
f1.close()
sigfile.close()
# Write delta file
f2 = open(dir2fn, "r")
sigfile = open("sig", "rb")
deltafile = open("delta", "wb")
librsync.filerdelta(sigfile, f2, deltafile)
f2.close()
sigfile.close()
deltafile.close()
# Write patched file
f1 = open(dir1fn, "rb")
newfile = open("%s/%s.out" % (dir1, i), "wb")
deltafile = open("delta", "rb")
librsync.filepatch(f1, deltafile, newfile)
f1.close()
deltafile.close()
newfile.close()
#!/usr/bin/env python
"""Make 10000 files consisting of data
Syntax: test.py directory_name number_of_files character filelength"""
import sys, os
dirname = sys.argv[1]
num_files = int(sys.argv[2])
character = sys.argv[3]
filelength = int(sys.argv[4])
os.mkdir(dirname)
for i in xrange(num_files):
fp = open("%s/%s" % (dirname, i), "w")
fp.write(character * filelength)
fp.close()
fp = open("%s.big" % dirname, "w")
fp.write(character * (filelength*num_files))
fp.close()
#!/usr/bin/python
import sys, os
sys.path.insert(0, "../rdiff_backup")
from rpath import *
from connection import *
import Globals
lc = Globals.local_connection
for filename in sys.argv[1:]:
#print "Deleting %s" % filename
rp = RPath(lc, filename)
if rp.lstat(): rp.delete()
#os.system("rm -rf " + rp.path)
#!/usr/bin/env python
"""Run rdiff to transform everything in one dir to another"""
import sys, os
dir1, dir2 = sys.argv[1:3]
for i in xrange(1000):
assert not os.system("rdiff signature %s/%s sig" % (dir1, i))
assert not os.system("rdiff delta sig %s/%s diff" % (dir2, i))
assert not os.system("rdiff patch %s/%s diff %s/%s.out" %
(dir1, i, dir1, i))
#!/usr/bin/env python
"""remove-comments.py
Given a python program on standard input, spit one out on stdout that
should work the same, but has blank and comment lines removed.
"""
import sys, re
triple_regex = re.compile('"""')
def eattriple(initial_line_stripped):
"""Keep reading until end of doc string"""
assert initial_line_stripped.startswith('"""')
if triple_regex.search(initial_line_stripped[3:]): return
while 1:
line = sys.stdin.readline()
if not line or triple_regex.search(line): break
while 1:
line = sys.stdin.readline()
if not line: break
stripped = line.strip()
if not stripped: continue
if stripped[0] == "#": continue
if stripped.startswith('"""'):
eattriple(stripped)
continue
sys.stdout.write(line)
#!/usr/bin/env python
# rdiff-backup -- Mirror files while keeping incremental changes
# Version $version released September 16, 2002
# Copyright (C) 2001, 2002 Ben Escoto <bescoto@stanford.edu>
#
# This program is licensed under the GNU General Public License (GPL).
# you can redistribute it and/or modify it under the terms of the GNU
# General Public License as published by the Free Software Foundation,
# Inc., 675 Mass Ave, Cambridge MA 02139, USA; either version 2 of the
# License, or (at your option) any later version. Distributions of
# rdiff-backup should include a copy of the GPL in a file called
# COPYING. The GPL is also available online at
# http://www.gnu.org/copyleft/gpl.html.
#
# See http://rdiff-backup.stanford.edu/ for more information. Please
# send mail to me or the mailing list if you find bugs or have any
# suggestions.
import sys
import rdiff_backup.Main
if __name__ == "__main__" and not globals().has_key('__no_execute__'):
rdiff_backup.Main.Main(sys.argv[1:])
This diff is collapsed.
# Copyright 2002 Ben Escoto
#
# This file is part of rdiff-backup.
#
# rdiff-backup 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, Inc., 675 Mass Ave, Cambridge MA
# 02139, USA; either version 2 of the License, or (at your option) any
# later version; incorporated herein by reference.
"""Coordinate corresponding files with different names
For instance, some source filenames may contain characters not allowed
on the mirror end. Also, if a source filename is very long (say 240
characters), the extra characters added to related increments may put
them over the usual 255 character limit.
"""
import re
from log import *
from robust import *
import Globals
max_filename_length = 255
# If true, enable character quoting, and set characters making
# regex-style range.
chars_to_quote = None
# These compiled regular expressions are used in quoting and unquoting
chars_to_quote_regexp = None
unquoting_regexp = None
# Use given char to quote. Default is set in Globals.
quoting_char = None
def set_init_quote_vals():
"""Set quoting value from Globals on all conns"""
for conn in Globals.connections:
conn.FilenameMapping.set_init_quote_vals_local()
def set_init_quote_vals_local():
"""Set value on local connection, initialize regexps"""
global chars_to_quote, quoting_char
chars_to_quote = Globals.chars_to_quote
if len(Globals.quoting_char) != 1:
Log.FatalError("Expected single character for quoting char,"
"got '%s' instead" % (Globals.quoting_char,))
quoting_char = Globals.quoting_char
init_quoting_regexps()
def init_quoting_regexps():
"""Compile quoting regular expressions"""
global chars_to_quote_regexp, unquoting_regexp
try:
chars_to_quote_regexp = \
re.compile("[%s%s]" % (chars_to_quote, quoting_char), re.S)
unquoting_regexp = re.compile("%s[0-9]{3}" % quoting_char, re.S)
except re.error:
Log.FatalError("Error '%s' when processing char quote list %s" %
(re.error, chars_to_quote))
def quote(path):
"""Return quoted version of given path
Any characters quoted will be replaced by the quoting char and
the ascii number of the character. For instance, "10:11:12"
would go to "10;05811;05812" if ":" were quoted and ";" were
the quoting character.
"""
return chars_to_quote_regexp.sub(quote_single, path)
def quote_single(match):
"""Return replacement for a single character"""
return "%s%03d" % (quoting_char, ord(match.group()))
def unquote(path):
"""Return original version of quoted filename"""
return unquoting_regexp.sub(unquote_single, path)
def unquote_single(match):
"""Unquote a single quoted character"""
assert len(match.group()) == 4
return chr(int(match.group()[1:]))
def get_quoted_dir_children(rpath):
"""For rpath directory, return list of quoted children in dir"""
if not rpath.isdir(): return []
dir_pairs = [(unquote(filename), filename)
for filename in Robust.listrp(rpath)]
dir_pairs.sort() # sort by real index, not quoted part
child_list = []
for unquoted, filename in dir_pairs:
childrp = rpath.append(unquoted)
childrp.quote_path()
child_list.append(childrp)
return child_list
# Copyright 2002 Ben Escoto
#
# This file is part of rdiff-backup.
#
# rdiff-backup 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, Inc., 675 Mass Ave, Cambridge MA
# 02139, USA; either version 2 of the License, or (at your option) any
# later version; incorporated herein by reference.
"""Hold a variety of constants usually set at initialization."""
import re, os
# The current version of rdiff-backup
version = "$version"
# If this is set, use this value in seconds as the current time
# instead of reading it from the clock.
current_time = None
# This determines how many bytes to read at a time when copying
blocksize = 32768
# This is used by the BufferedRead class to determine how many
# bytes to request from the underlying file per read(). Larger
# values may save on connection overhead and latency.
conn_bufsize = 98304
# True if script is running as a server
server = None
# uid and gid of the owner of the rdiff-backup process. This can
# vary depending on the connection.
process_uid = os.getuid()
process_gid = os.getgid()
# If true, when copying attributes, also change target's uid/gid
change_ownership = None
# If true, change the permissions of unwriteable mirror files
# (such as directories) so that they can be written, and then
# change them back. This defaults to 1 just in case the process
# is not running as root (root doesn't need to change
# permissions).
change_mirror_perms = (process_uid != 0)
# If true, temporarily change permissions of unreadable files in
# the source directory to make sure we can read all files.
change_source_perms = None
# If true, try to reset the atimes of the source partition.
preserve_atime = None
# This will be set as soon as the LocalConnection class loads
local_connection = None
# All connections should be added to the following list, so
# further global changes can be propagated to the remote systems.
# The first element should be Globals.local_connection. For a
# server, the second is the connection to the client.
connections = []
# Each process should have a connection number unique to the
# session. The client has connection number 0.
connection_number = 0
# Dictionary pairing connection numbers with connections. Set in
# SetConnections for all connections.
connection_dict = {}
# True if the script is the end that reads the source directory
# for backups. It is true for purely local sessions.
isbackup_reader = None
# Connection of the real backup reader (for which isbackup_reader
# is true)
backup_reader = None
# True if the script is the end that writes to the increment and
# mirror directories. True for purely local sessions.
isbackup_writer = None
# Connection of the backup writer
backup_writer = None
# Connection of the client
client_conn = None
# This list is used by the set function below. When a new
# connection is created with init_connection, its Globals class
# will match this one for all the variables mentioned in this
# list.
changed_settings = []
# rdiff-backup will try to checkpoint its state every
# checkpoint_interval seconds. Then when resuming, at most this
# amount of time is lost.
checkpoint_interval = 20
# The RPath of the rdiff-backup-data directory.
rbdir = None
# Indicates if a resume or a lack of resume is forced. This
# should be None for the default. 0 means don't resume, and 1
# means resume.
resume = None
# If there has been an aborted backup fewer than this many seconds
# ago, attempt to resume it where it left off instead of starting
# a new one.
resume_window = 7200
# This string is used when recognizing and creating time strings.
# If the time_separator is ":", then W3 datetime strings like
# 2001-12-07T04:22:01-07:00 are produced. It can be set to "_" to
# make filenames that don't contain colons, which aren't allowed
# under MS windows NT.
time_separator = ":"
# quoting_enabled is true if we should quote certain characters in
# filenames on the source side (see FilenameMapping for more
# info). chars_to_quote is a string whose characters should be
# quoted, and quoting_char is the character to quote with.
quoting_enabled = None
chars_to_quote = ""
quoting_char = ';'
# If true, emit output intended to be easily readable by a
# computer. False means output is intended for humans.
parsable_output = None
# If true, then hardlinks will be preserved to mirror and recorded
# in the increments directory. There is also a difference here
# between None and 0. When restoring, None or 1 means to preserve
# hardlinks iff can find a hardlink dictionary. 0 means ignore
# hardlink information regardless.
preserve_hardlinks = 1
# If this is false, then rdiff-backup will not compress any
# increments. Default is to compress based on regexp below.
compression = 1
# Increments based on files whose names match this
# case-insensitive regular expression won't be compressed (applies
# to .snapshots and .diffs). The second below will be the
# compiled version of the first.
no_compression_regexp_string = "(?i).*\\.(gz|z|bz|bz2|tgz|zip|rpm|deb|" \
"jpg|gif|png|jp2|mp3|ogg|avi|wmv|mpeg|mpg|rm|mov)$"
no_compression_regexp = None
# If true, filelists and directory statistics will be split on
# nulls instead of newlines.
null_separator = None
# Determines whether or not ssh will be run with the -C switch
ssh_compression = 1
# If true, print statistics after successful backup
print_statistics = None
# On the reader and writer connections, the following will be
# replaced by the source and mirror Select objects respectively.
select_source, select_mirror = None, None
# On the backup writer connection, holds the root incrementing branch
# object. Access is provided to increment error counts.
ITRB = None
# Percentage of time to spend sleeping. None means never sleep.
sleep_ratio = None
# security_level has 4 values and controls which requests from remote
# systems will be honored. "all" means anything goes. "read-only"
# means that the requests must not write to disk. "update-only" means
# that requests shouldn't destructively update the disk (but normal
# incremental updates are OK). "minimal" means only listen to a few
# basic requests.
security_level = "all"
# If this is set, it indicates that the remote connection should only
# deal with paths inside of restrict_path.
restrict_path = None
def get(name):
"""Return the value of something in this module"""
return globals()[name]
def is_not_None(name):
"""Returns true if value is not None"""
return globals()[name] is not None
def set(name, val):
"""Set the value of something in this module
Use this instead of writing the values directly if the setting
matters to remote sides. This function updates the
changed_settings list, so other connections know to copy the
changes.
"""
changed_settings.append(name)
globals()[name] = val
def set_integer(name, val):
"""Like set, but make sure val is an integer"""
try: intval = int(val)
except ValueError:
Log.FatalError("Variable %s must be set to an integer -\n"
"received %s instead." % (name, val))
set(name, intval)
def set_float(name, val, min = None, max = None, inclusive = 1):
"""Like set, but make sure val is float within given bounds"""
def error():
s = "Variable %s must be set to a float" % (name,)
if min is not None and max is not None:
s += " between %s and %s " % (min, max)
if inclusive: s += "inclusive"
else: s += "not inclusive"
elif min is not None or max is not None:
if inclusive: inclusive_string = "or equal to "
else: inclusive_string = ""
if min is not None:
s += " greater than %s%s" % (inclusive_string, min)
else: s+= " less than %s%s" % (inclusive_string, max)
Log.FatalError(s)
try: f = float(val)
except ValueError: error()
if min is not None:
if inclusive and f < min: error()
elif not inclusive and f <= min: error()
if max is not None:
if inclusive and f > max: error()
elif not inclusive and f >= max: error()
set(name, f)
def get_dict_val(name, key):
"""Return val from dictionary in this class"""
return globals()[name][key]
def set_dict_val(name, key, val):
"""Set value for dictionary in this class"""
globals()[name][key] = val
def postset_regexp(name, re_string, flags = None):
"""Compile re_string on all existing connections, set to name"""
for conn in connections:
conn.Globals.postset_regexp_local(name, re_string, flags)
def postset_regexp_local(name, re_string, flags):
"""Set name to compiled re_string locally"""
if flags: globals()[name] = re.compile(re_string, flags)
else: globals()[name] = re.compile(re_string)
def set_select(dsrpath, tuplelist, quote_mode, *filelists):
"""Initialize select object using tuplelist
Note that each list in filelists must each be passed as
separate arguments, so each is recognized as a file by the
connection. Otherwise we will get an error because a list
containing files can't be pickled.
"""
global select_source, select_mirror
if dsrpath.source:
select_source = Select(dsrpath, quote_mode)
select_source.ParseArgs(tuplelist, filelists)
else:
select_mirror = Select(dsrpath, quote_mode)
select_mirror.ParseArgs(tuplelist, filelists)
from rpath import * # kludge to avoid circularity - not needed in this module
from selection import *
# Copyright 2002 Ben Escoto
#
# This file is part of rdiff-backup.
#
# rdiff-backup 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, Inc., 675 Mass Ave, Cambridge MA
# 02139, USA; either version 2 of the License, or (at your option) any
# later version; incorporated herein by reference.
"""Preserve and restore hard links
If the preserve_hardlinks option is selected, linked files in the
source directory will be linked in the mirror directory. Linked files
are treated like any other with respect to incrementing, but a
database of all links will be recorded at each session, so linked
files can still be restored from the increments.
All these functions are meant to be executed on the destination
side. The source side should only transmit inode information.
"""
from __future__ import generators
import cPickle
# In all of these lists of indicies are the values. The keys in
# _inode_ ones are (inode, devloc) pairs.
_src_inode_indicies = {}
_dest_inode_indicies = {}
# The keys for these two are just indicies. They share values
# with the earlier dictionaries.
_src_index_indicies = {}
_dest_index_indicies = {}
# When a linked file is restored, its path is added to this dict,
# so it can be found when later paths being restored are linked to
# it.
_restore_index_path = {}
def get_inode_key(rorp):
"""Return rorp's key for _inode_ dictionaries"""
return (rorp.getinode(), rorp.getdevloc())
def get_indicies(rorp, source):
"""Return a list of similarly linked indicies, using rorp's index"""
if source: dict = _src_index_indicies
else: dict = _dest_index_indicies
try: return dict[rorp.index]
except KeyError: return []
def add_rorp(rorp, source):
"""Process new rorp and update hard link dictionaries
First enter it into src_inode_indicies. If we have already
seen all the hard links, then we can delete the entry.
Everything must stay recorded in src_index_indicies though.
"""
if not rorp.isreg() or rorp.getnumlinks() < 2: return
if source:
inode_dict, index_dict = _src_inode_indicies, _src_index_indicies
else: inode_dict, index_dict = _dest_inode_indicies, _dest_index_indicies
rp_inode_key = get_inode_key(rorp)
if inode_dict.has_key(rp_inode_key):
index_list = inode_dict[rp_inode_key]
index_list.append(rorp.index)
if len(index_list) == rorp.getnumlinks():
del inode_dict[rp_inode_key]
else: # make new entry in both src dicts
index_list = [rorp.index]
inode_dict[rp_inode_key] = index_list
index_dict[rorp.index] = index_list
def add_rorp_iter(iter, source):
"""Return new rorp iterator like iter that add_rorp's first"""
for rorp in iter:
add_rorp(rorp, source)
yield rorp
def rorp_eq(src_rorp, dest_rorp):
"""Compare hardlinked for equality
Two files may otherwise seem equal but be hardlinked in
different ways. This function considers them equal enough if
they have been hardlinked correctly to the previously seen
indicies.
"""
if not src_rorp.index == dest_rorp.index: return None
if (not src_rorp.isreg() or not dest_rorp.isreg() or
src_rorp.getnumlinks() == dest_rorp.getnumlinks() == 1):
return 1 # Hard links don't apply
src_index_list = get_indicies(src_rorp, 1)
dest_index_list = get_indicies(dest_rorp, None)
# If a list only has one element, then it is only hardlinked
# to itself so far, so that is not a genuine difference yet.
if not src_index_list or len(src_index_list) == 1:
return not dest_index_list or len(dest_index_list) == 1
if not dest_index_list or len(dest_index_list) == 1: return None
# Both index lists exist and are non-empty
return src_index_list == dest_index_list # they are always sorted
def islinked(rorp):
"""True if rorp's index is already linked to something on src side"""
return len(get_indicies(rorp, 1)) >= 2
def restore_link(index, rpath):
"""Restores a linked file by linking it
When restoring, all the hardlink data is already present, and
we can only link to something already written. In either
case, add to the _restore_index_path dict, so we know later
that the file is available for hard
linking.
Returns true if succeeded in creating rpath, false if must
restore rpath normally.
"""
if index not in _src_index_indicies: return None
for linked_index in _src_index_indicies[index]:
if linked_index in _restore_index_path:
srcpath = _restore_index_path[linked_index]
Log("Restoring %s by hard linking to %s" %
(rpath.path, srcpath), 6)
rpath.hardlink(srcpath)
return 1
_restore_index_path[index] = rpath.path
return None
def link_rp(src_rorp, dest_rpath, dest_root = None):
"""Make dest_rpath into a link analogous to that of src_rorp"""
if not dest_root: dest_root = dest_rpath # use base of dest_rpath
dest_link_rpath = RPath(dest_root.conn, dest_root.base,
get_indicies(src_rorp, 1)[0])
dest_rpath.hardlink(dest_link_rpath.path)
def write_linkdict(rpath, dict, compress = None):
"""Write link data to the rbdata dir
It is stored as the a big pickled dictionary dated to match
the current hardlinks.
"""
assert (Globals.isbackup_writer and
rpath.conn is Globals.local_connection)
tf = TempFileManager.new(rpath)
def init():
fp = tf.open("wb", compress)
cPickle.dump(dict, fp)
assert not fp.close()
tf.setdata()
Robust.make_tf_robustaction(init, (tf,), (rpath,)).execute()
def get_linkrp(data_rpath, time, prefix):
"""Return RPath of linkdata, or None if cannot find"""
for rp in map(data_rpath.append, data_rpath.listdir()):
if (rp.isincfile() and rp.getincbase_str() == prefix and
(rp.getinctype() == 'snapshot' or rp.getinctype() == 'data')
and Time.stringtotime(rp.getinctime()) == time):
return rp
return None
def get_linkdata(data_rpath, time, prefix = 'hardlink_data'):
"""Return index dictionary written by write_linkdata at time"""
rp = get_linkrp(data_rpath, time, prefix)
if not rp: return None
fp = rp.open("rb", rp.isinccompressed())
index_dict = cPickle.load(fp)
assert not fp.close()
return index_dict
def final_writedata():
"""Write final checkpoint data to rbdir after successful backup"""
global final_inc
if _src_index_indicies:
Log("Writing hard link data", 6)
if Globals.compression:
final_inc = Globals.rbdir.append("hardlink_data.%s.data.gz" %
Time.curtimestr)
else: final_inc = Globals.rbdir.append("hardlink_data.%s.data" %
Time.curtimestr)
write_linkdict(final_inc, _src_index_indicies, Globals.compression)
else: # no hardlinks, so writing unnecessary
final_inc = None
def retrieve_final(time):
"""Set source index dictionary from hardlink_data file if avail"""
global _src_index_indicies
hd = get_linkdata(Globals.rbdir, time)
if hd is None: return None
_src_index_indicies = hd
return 1
def final_checkpoint(data_rpath):
"""Write contents of the four dictionaries to the data dir
If rdiff-backup receives a fatal error, it may still be able
to save the contents of the four hard link dictionaries.
Because these dictionaries may be big, they are not saved
after every 20 seconds or whatever, but just at the end.
"""
Log("Writing intermediate hard link data to disk", 2)
src_inode_rp = data_rpath.append("hardlink_source_inode_checkpoint."
"%s.data" % Time.curtimestr)
src_index_rp = data_rpath.append("hardlink_source_index_checkpoint."
"%s.data" % Time.curtimestr)
dest_inode_rp = data_rpath.append("hardlink_dest_inode_checkpoint."
"%s.data" % Time.curtimestr)
dest_index_rp = data_rpath.append("hardlink_dest_index_checkpoint."
"%s.data" % Time.curtimestr)
for (rp, dict) in ((src_inode_rp, _src_inode_indicies),
(src_index_rp, _src_index_indicies),
(dest_inode_rp, _dest_inode_indicies),
(dest_index_rp, _dest_index_indicies)):
write_linkdict(rp, dict)
def retrieve_checkpoint(data_rpath, time):
"""Retrieve hardlink data from final checkpoint
Return true if the retrieval worked, false otherwise.
"""
global _src_inode_indicies, _src_index_indicies
global _dest_inode_indicies, _dest_index_indicies
try:
src_inode = get_linkdata(data_rpath, time,
"hardlink_source_inode_checkpoint")
src_index = get_linkdata(data_rpath, time,
"hardlink_source_index_checkpoint")
dest_inode = get_linkdata(data_rpath, time,
"hardlink_dest_inode_checkpoint")
dest_index = get_linkdata(data_rpath, time,
"hardlink_dest_index_checkpoint")
except cPickle.UnpicklingError:
Log("Unpickling Error", 2)
return None
if (src_inode is None or src_index is None or
dest_inode is None or dest_index is None): return None
_src_inode_indicies, _src_index_indicies = src_inode, src_index
_dest_inode_indicies, _dest_index_indicies = dest_inode, dest_index
return 1
def remove_all_checkpoints():
"""Remove all hardlink checkpoint information from directory"""
prefix_list = ["hardlink_source_inode_checkpoint",
"hardlink_source_index_checkpoint",
"hardlink_dest_inode_checkpoint",
"hardlink_dest_index_checkpoint"]
for rp in map(Globals.rbdir.append, Globals.rbdir.listdir()):
if (rp.isincfile() and rp.getincbase_str() in prefix_list and
(rp.getinctype() == 'snapshot' or rp.getinctype() == 'data')):
rp.delete()
from log import *
from robust import *
from rpath import *
import Globals, Time
This diff is collapsed.
#!/usr/bin/env python
"""Read component files of rdiff-backup, and glue them together after
removing unnecessary bits."""
import os
def mystrip(filename):
"""Open filename, read input, strip appropriately, and return contents"""
fp = open(filename, "r")
lines = fp.readlines()
fp.close()
i = 0
while(lines[i][:60] !=
"############################################################"):
i = i+1
return "".join(lines[i:]).strip() + "\n\n\n"
files = ["globals.py", "static.py", "lazy.py", "log.py", "ttime.py",
"iterfile.py", "rdiff.py", "connection.py", "rpath.py",
"hardlink.py", "robust.py", "rorpiter.py",
"destructive_stepping.py", "selection.py",
"filename_mapping.py", "statistics.py", "increment.py",
"restore.py", "manage.py", "highlevel.py",
"setconnections.py", "main.py"]
os.system("cp header.py rdiff-backup")
outfp = open("rdiff-backup", "a")
for file in files:
outfp.write(mystrip(file))
outfp.close()
os.system("chmod 755 rdiff-backup")
# Copyright 2002 Ben Escoto
#
# This file is part of rdiff-backup.
#
# rdiff-backup 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, Inc., 675 Mass Ave, Cambridge MA
# 02139, USA; either version 2 of the License, or (at your option) any
# later version; incorporated herein by reference.
"""Misc statistics methods, pertaining to dir and session stat files"""
from statistics import *
# This is the RPath of the directory statistics file, and the
# associated open file. It will hold a line of statistics for
# each directory that is backed up.
_dir_stats_rp = None
_dir_stats_fp = None
# This goes at the beginning of the directory statistics file and
# explains the format.
_dir_stats_header = """# rdiff-backup directory statistics file
#
# Each line is in the following format:
# RelativeDirName %s
""" % " ".join(StatsObj.stat_file_attrs)
def open_dir_stats_file():
"""Open directory statistics file, write header"""
global _dir_stats_fp, _dir_stats_rp
assert not _dir_stats_fp, "Directory file already open"
if Globals.compression: suffix = "data.gz"
else: suffix = "data"
_dir_stats_rp = Inc.get_inc(Globals.rbdir.append("directory_statistics"),
Time.curtime, suffix)
if _dir_stats_rp.lstat():
Log("Warning, statistics file %s already exists, appending" %
_dir_stats_rp.path, 2)
_dir_stats_fp = _dir_stats_rp.open("ab", Globals.compression)
else: _dir_stats_fp = _dir_stats_rp.open("wb", Globals.compression)
_dir_stats_fp.write(_dir_stats_header)
def write_dir_stats_line(statobj, index):
"""Write info from statobj about rpath to statistics file"""
if Globals.null_separator:
_dir_stats_fp.write(statobj.get_stats_line(index, None) + "\0")
else: _dir_stats_fp.write(statobj.get_stats_line(index) + "\n")
def close_dir_stats_file():
"""Close directory statistics file if its open"""
global _dir_stats_fp
if _dir_stats_fp:
_dir_stats_fp.close()
_dir_stats_fp = None
def write_session_statistics(statobj):
"""Write session statistics into file, log"""
stat_inc = Inc.get_inc(Globals.rbdir.append("session_statistics"),
Time.curtime, "data")
statobj.StartTime = Time.curtime
statobj.EndTime = time.time()
# include hardlink data and dir stats in size of increments
if Globals.preserve_hardlinks and Hardlink.final_inc:
# include hardlink data in size of increments
statobj.IncrementFiles += 1
statobj.IncrementFileSize += Hardlink.final_inc.getsize()
if _dir_stats_rp and _dir_stats_rp.lstat():
statobj.IncrementFiles += 1
statobj.IncrementFileSize += _dir_stats_rp.getsize()
statobj.write_stats_to_rp(stat_inc)
if Globals.print_statistics:
message = statobj.get_stats_logstring("Session statistics")
Log.log_to_file(message)
Globals.client_conn.sys.stdout.write(message)
from increment import *
import Hardlink
# Copyright 2002 Ben Escoto
#
# This file is part of rdiff-backup.
#
# rdiff-backup 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, Inc., 675 Mass Ave, Cambridge MA
# 02139, USA; either version 2 of the License, or (at your option) any
# later version; incorporated herein by reference.
"""Invoke rdiff utility to make signatures, deltas, or patch
All these operations should be done in a relatively safe manner using
RobustAction and the like.
"""
import os, librsync
class RdiffException(Exception): pass
def get_signature(rp):
"""Take signature of rpin file and return in file object"""
Log("Getting signature of %s" % rp.path, 7)
return librsync.SigFile(rp.open("rb"))
def get_delta_sigfileobj(sig_fileobj, rp_new):
"""Like get_delta but signature is in a file object"""
Log("Getting delta of %s with signature stream" % (rp_new.path,), 7)
return librsync.DeltaFile(sig_fileobj, rp_new.open("rb"))
def get_delta_sigrp(rp_signature, rp_new):
"""Take signature rp and new rp, return delta file object"""
Log("Getting delta of %s with signature %s" %
(rp_new.path, rp_signature.path), 7)
return librsync.DeltaFile(rp_signature.open("rb"), rp_new.open("rb"))
def write_delta_action(basis, new, delta, compress = None):
"""Return action writing delta which brings basis to new
If compress is true, the output of rdiff will be gzipped
before written to delta.
"""
delta_tf = TempFileManager.new(delta)
def init(): write_delta(basis, new, delta_tf, compress)
return Robust.make_tf_robustaction(init, delta_tf, delta)
def write_delta(basis, new, delta, compress = None):
"""Write rdiff delta which brings basis to new"""
Log("Writing delta %s from %s -> %s" %
(basis.path, new.path, delta.path), 7)
sigfile = librsync.SigFile(basis.open("rb"))
deltafile = librsync.DeltaFile(sigfile, new.open("rb"))
delta.write_from_fileobj(deltafile, compress)
def patch_action(rp_basis, rp_delta, rp_out = None, out_tf = None,
delta_compressed = None):
"""Return RobustAction which patches rp_basis with rp_delta
If rp_out is None, put output in rp_basis. Will use TempFile
out_tf it is specified. If delta_compressed is true, the
delta file will be decompressed before processing with rdiff.
"""
if not rp_out: rp_out = rp_basis
if not out_tf: out_tf = TempFileManager.new(rp_out)
def init():
rp_basis.conn.Rdiff.patch_local(rp_basis, rp_delta,
out_tf, delta_compressed)
out_tf.setdata()
return Robust.make_tf_robustaction(init, out_tf, rp_out)
def patch_local(rp_basis, rp_delta, outrp, delta_compressed = None):
"""Patch routine that must be run on rp_basis.conn
This is because librsync may need to seek() around in rp_basis,
and so needs a real file. Other rpaths can be remote.
"""
assert rp_basis.conn is Globals.local_connection
if delta_compressed: deltafile = rp_delta.open("rb", 1)
else: deltafile = rp_delta.open("rb")
sigfile = librsync.SigFile(rp_basis.open("rb"))
patchfile = librsync.PatchedFile(rp_basis.open("rb"), deltafile)
outrp.write_from_fileobj(patchfile)
def patch_with_attribs_action(rp_basis, rp_delta, rp_out = None):
"""Like patch_action, but also transfers attributs from rp_delta"""
if not rp_out: rp_out = rp_basis
tf = TempFileManager.new(rp_out)
return Robust.chain_nested(patch_action(rp_basis, rp_delta, rp_out, tf),
Robust.copy_attribs_action(rp_delta, tf))
def copy_action(rpin, rpout):
"""Use rdiff to copy rpin to rpout, conserving bandwidth"""
if not rpin.isreg() or not rpout.isreg() or rpin.conn is rpout.conn:
# rdiff not applicable, fallback to regular copying
return Robust.copy_action(rpin, rpout)
Log("Rdiff copying %s to %s" % (rpin.path, rpout.path), 6)
out_tf = TempFileManager.new(rpout)
def init(): rpout.conn.Rdiff.copy_local(rpin, rpout, out_tf)
return Robust.make_tf_robustaction(init, out_tf, rpout)
def copy_local(rpin, rpout, rpnew):
"""Write rpnew == rpin using rpout as basis. rpout and rpnew local"""
assert rpnew.conn is rpout.conn is Globals.local_connection
sigfile = librsync.SigFile(rpout.open("rb"))
deltafile = rpin.conn.librsync.DeltaFile(sigfile, rpin.open("rb"))
rpnew.write_from_fileobj(librsync.PatchedFile(rpout.open("rb"), deltafile))
from log import *
from robust import *
# Copyright 2002 Ben Escoto
#
# This file is part of rdiff-backup.
#
# rdiff-backup 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, Inc., 675 Mass Ave, Cambridge MA
# 02139, USA; either version 2 of the License, or (at your option) any
# later version; incorporated herein by reference.
"""Functions to make sure remote requests are kosher"""
import sys, tempfile
import Globals, Main
from rpath import *
class Violation(Exception):
"""Exception that indicates an improper request has been received"""
pass
# This will store the list of functions that will be honored from
# remote connections.
allowed_requests = None
# This stores the list of global variables that the client can not
# set on the server.
disallowed_server_globals = ["server", "security_level", "restrict_path"]
def initialize(action, cmdpairs):
"""Initialize allowable request list and chroot"""
global allowed_requests
set_security_level(action, cmdpairs)
set_allowed_requests(Globals.security_level)
def set_security_level(action, cmdpairs):
"""If running client, set security level and restrict_path
To find these settings, we must look at the action to see what is
supposed to happen, and then look at the cmdpairs to see what end
the client is on.
"""
def islocal(cmdpair): return not cmdpair[0]
def bothlocal(cp1, cp2): return islocal(cp1) and islocal(cp2)
def bothremote(cp1, cp2): return not islocal(cp1) and not islocal(cp2)
def getpath(cmdpair): return cmdpair[1]
if Globals.server: return
cp1 = cmdpairs[0]
if len(cmdpairs) > 1: cp2 = cmdpairs[1]
if action == "backup":
if bothlocal(cp1, cp2) or bothremote(cp1, cp2):
sec_level = "minimal"
rdir = tempfile.gettempdir()
elif islocal(cp1):
sec_level = "read-only"
rdir = getpath(cp1)
else:
assert islocal(cp2)
sec_level = "update-only"
rdir = getpath(cp2)
elif action == "restore" or action == "restore-as-of":
if len(cmdpairs) == 1 or bothlocal(cp1, cp2) or bothremote(cp1, cp2):
sec_level = "minimal"
rdir = tempfile.gettempdir()
elif islocal(cp1):
sec_level = "read-only"
rdir = Main.restore_get_root(RPath(Globals.local_connection,
getpath(cp1)))[0].path
else:
assert islocal(cp2)
sec_level = "all"
rdir = getpath(cp2)
elif action == "mirror":
if bothlocal(cp1, cp2) or bothremote(cp1, cp2):
sec_level = "minimal"
rdir = tempfile.gettempdir()
elif islocal(cp1):
sec_level = "read-only"
rdir = getpath(cp1)
else:
assert islocal(cp2)
sec_level = "all"
rdir = getpath(cp2)
elif (action == "test-server" or action == "list-increments" or
action == "calculate-average" or action == "remove-older-than"):
sec_level = "minimal"
rdir = tempfile.gettempdir()
else: assert 0, "Unknown action %s" % action
Globals.security_level = sec_level
Globals.restrict_path = RPath(Globals.local_connection,
rdir).normalize().path
def set_allowed_requests(sec_level):
"""Set the allowed requests list using the security level"""
global allowed_requests
if sec_level == "all": return
allowed_requests = ["VirtualFile.readfromid", "VirtualFile.closebyid",
"Globals.get", "Globals.is_not_None",
"Globals.get_dict_val",
"Log.open_logfile_allconn",
"Log.close_logfile_allconn",
"SetConnections.add_redirected_conn",
"RedirectedRun",
"sys.stdout.write"]
if sec_level == "minimal": pass
elif sec_level == "read-only" or sec_level == "update-only":
allowed_requests.extend(["C.make_file_dict",
"os.getuid",
"os.listdir",
"Time.setcurtime_local",
"Resume.ResumeCheck",
"HLSourceStruct.split_initial_dsiter",
"HLSourceStruct.get_diffs_and_finalize",
"RPathStatic.gzip_open_local_read",
"RPathStatic.open_local_read"])
if sec_level == "update-only":
allowed_requests. \
extend(["Log.open_logfile_local", "Log.close_logfile_local",
"Log.close_logfile_allconn", "Log.log_to_file",
"SaveState.init_filenames",
"SaveState.touch_last_file",
"HLDestinationStruct.get_sigs",
"HLDestinationStruct.patch_w_datadir_writes",
"HLDestinationStruct.patch_and_finalize",
"HLDestinationStruct.patch_increment_and_finalize",
"Main.backup_touch_curmirror_local",
"Globals.ITRB.increment_stat"])
if Globals.server:
allowed_requests.extend(["SetConnections.init_connection_remote",
"Log.setverbosity",
"Log.setterm_verbosity",
"Time.setprevtime_local",
"FilenameMapping.set_init_quote_vals_local",
"Globals.postset_regexp_local",
"Globals.set_select",
"HLSourceStruct.set_session_info",
"HLDestinationStruct.set_session_info"])
def vet_request(request, arglist):
"""Examine request for security violations"""
#if Globals.server: sys.stderr.write(str(request) + "\n")
security_level = Globals.security_level
if Globals.restrict_path:
for arg in arglist:
if isinstance(arg, RPath): vet_rpath(arg)
if security_level == "all": return
if request.function_string in allowed_requests: return
if request.function_string == "Globals.set":
if Globals.server and arglist[0] not in disallowed_server_globals:
return
raise Violation("\nWarning Security Violation!\n"
"Bad request for function: %s\n"
"with arguments: %s\n" % (request.function_string,
arglist))
def vet_rpath(rpath):
"""Require rpath not to step outside retricted directory"""
if Globals.restrict_path and rpath.conn is Globals.local_connection:
normalized, restrict = rpath.normalize().path, Globals.restrict_path
components = normalized.split("/")
# 3 cases for restricted dir /usr/foo: /var, /usr/foobar, /usr/foo/..
if (not normalized.startswith(restrict) or
(len(normalized) > len(restrict) and
normalized[len(restrict)] != "/") or
".." in components):
raise Violation("\nWarning Security Violation!\n"
"Request to handle path %s\n"
"which doesn't appear to be within "
"restrict path %s.\n" % (normalized, restrict))
# Copyright 2002 Ben Escoto
#
# This file is part of rdiff-backup.
#
# rdiff-backup 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, Inc., 675 Mass Ave, Cambridge MA
# 02139, USA; either version 2 of the License, or (at your option) any
# later version; incorporated herein by reference.
"""Parse args and setup connections
The functions in this module are used once by Main to parse file
descriptions like bescoto@folly.stanford.edu:/usr/bin/ls and to set up
the related connections.
"""
# This is the schema that determines how rdiff-backup will open a
# pipe to the remote system. If the file is given as A::B, %s will
# be substituted with A in the schema.
__cmd_schema = 'ssh -C %s rdiff-backup --server'
__cmd_schema_no_compress = 'ssh %s rdiff-backup --server'
# This is a list of remote commands used to start the connections.
# The first is None because it is the local connection.
__conn_remote_cmds = [None]
class SetConnectionsException(Exception): pass
def get_cmd_pairs(arglist, remote_schema = None, remote_cmd = None):
"""Map the given file descriptions into command pairs
Command pairs are tuples cmdpair with length 2. cmdpair[0] is
None iff it describes a local path, and cmdpair[1] is the path.
"""
global __cmd_schema
if remote_schema: __cmd_schema = remote_schema
elif not Globals.ssh_compression: __cmd_schema = __cmd_schema_no_compress
if not arglist: return []
desc_pairs = map(parse_file_desc, arglist)
if filter(lambda x: x[0], desc_pairs): # True if any host_info found
if remote_cmd:
Log.FatalError("The --remote-cmd flag is not compatible "
"with remote file descriptions.")
elif remote_schema:
Log("Remote schema option ignored - no remote file "
"descriptions.", 2)
cmdpairs = map(desc2cmd_pairs, desc_pairs)
if remote_cmd: # last file description gets remote_cmd
cmd_pairs[-1] = (remote_cmd, cmd_pairs[-1][1])
return cmdpairs
def cmdpair2rp(cmd_pair):
"""Return RPath from cmd_pair (remote_cmd, filename)"""
cmd, filename = cmd_pair
if cmd: conn = init_connection(cmd)
else: conn = Globals.local_connection
return RPath(conn, filename)
def desc2cmd_pairs(desc_pair):
"""Return pair (remote_cmd, filename) from desc_pair"""
host_info, filename = desc_pair
if not host_info: return (None, filename)
else: return (fill_schema(host_info), filename)
def parse_file_desc(file_desc):
"""Parse file description returning pair (host_info, filename)
In other words, bescoto@folly.stanford.edu::/usr/bin/ls =>
("bescoto@folly.stanford.edu", "/usr/bin/ls"). The
complication is to allow for quoting of : by a \. If the
string is not separated by :, then the host_info is None.
"""
def check_len(i):
if i >= len(file_desc):
raise SetConnectionsException(
"Unexpected end to file description %s" % file_desc)
host_info_list, i, last_was_quoted = [], 0, None
while 1:
if i == len(file_desc):
return (None, file_desc)
if file_desc[i] == '\\':
i = i+1
check_len(i)
last_was_quoted = 1
elif (file_desc[i] == ":" and i > 0 and file_desc[i-1] == ":"
and not last_was_quoted):
host_info_list.pop() # Remove last colon from name
break
else: last_was_quoted = None
host_info_list.append(file_desc[i])
i = i+1
check_len(i+1)
return ("".join(host_info_list), file_desc[i+1:])
def fill_schema(host_info):
"""Fills host_info into the schema and returns remote command"""
return __cmd_schema % host_info
def init_connection(remote_cmd):
"""Run remote_cmd, register connection, and then return it
If remote_cmd is None, then the local connection will be
returned. This also updates some settings on the remote side,
like global settings, its connection number, and verbosity.
"""
if not remote_cmd: return Globals.local_connection
Log("Executing " + remote_cmd, 4)
stdin, stdout = os.popen2(remote_cmd)
conn_number = len(Globals.connections)
conn = PipeConnection(stdout, stdin, conn_number)
check_connection_version(conn, remote_cmd)
Log("Registering connection %d" % conn_number, 7)
init_connection_routing(conn, conn_number, remote_cmd)
init_connection_settings(conn)
return conn
def check_connection_version(conn, remote_cmd):
"""Log warning if connection has different version"""
try: remote_version = conn.Globals.get('version')
except ConnectionReadError, exception:
Log.FatalError("""%s
Couldn't start up the remote connection by executing
%s
Remember that, under the default settings, rdiff-backup must be
installed in the PATH on the remote system. See the man page for more
information.""" % (exception, remote_cmd))
if remote_version != Globals.version:
Log("Warning: Local version %s does not match remote version %s."
% (Globals.version, remote_version), 2)
def init_connection_routing(conn, conn_number, remote_cmd):
"""Called by init_connection, establish routing, conn dict"""
Globals.connection_dict[conn_number] = conn
conn.SetConnections.init_connection_remote(conn_number)
for other_remote_conn in Globals.connections[1:]:
conn.SetConnections.add_redirected_conn(
other_remote_conn.conn_number)
other_remote_conn.SetConnections.add_redirected_conn(conn_number)
Globals.connections.append(conn)
__conn_remote_cmds.append(remote_cmd)
def init_connection_settings(conn):
"""Tell new conn about log settings and updated globals"""
conn.Log.setverbosity(Log.verbosity)
conn.Log.setterm_verbosity(Log.term_verbosity)
for setting_name in Globals.changed_settings:
conn.Globals.set(setting_name, Globals.get(setting_name))
FilenameMapping.set_init_quote_vals()
def init_connection_remote(conn_number):
"""Run on server side to tell self that have given conn_number"""
Globals.connection_number = conn_number
Globals.local_connection.conn_number = conn_number
Globals.connection_dict[0] = Globals.connections[1]
Globals.connection_dict[conn_number] = Globals.local_connection
def add_redirected_conn(conn_number):
"""Run on server side - tell about redirected connection"""
Globals.connection_dict[conn_number] = \
RedirectedConnection(conn_number)
def UpdateGlobal(setting_name, val):
"""Update value of global variable across all connections"""
for conn in Globals.connections:
conn.Globals.set(setting_name, val)
def BackupInitConnections(reading_conn, writing_conn):
"""Backup specific connection initialization"""
reading_conn.Globals.set("isbackup_reader", 1)
writing_conn.Globals.set("isbackup_writer", 1)
UpdateGlobal("backup_reader", reading_conn)
UpdateGlobal("backup_writer", writing_conn)
if (Globals.change_source_perms and
reading_conn.Globals.get("process_uid") == 0):
Log("Warning: --change_source_perms should usually not be used when\n"
"the reading connection is running as root, because root can\n"
"read all files regardless of their permissions.", 2)
def CloseConnections():
"""Close all connections. Run by client"""
assert not Globals.server
for conn in Globals.connections: conn.quit()
del Globals.connections[1:] # Only leave local connection
Globals.connection_dict = {0: Globals.local_connection}
Globals.backup_reader = Globals.isbackup_reader = \
Globals.backup_writer = Globals.isbackup_writer = None
def TestConnections():
"""Test connections, printing results"""
if len(Globals.connections) == 1: print "No remote connections specified"
else:
for i in range(1, len(Globals.connections)): test_connection(i)
def test_connection(conn_number):
"""Test connection. conn_number 0 is the local connection"""
print "Testing server started by: ", __conn_remote_cmds[conn_number]
conn = Globals.connections[conn_number]
try:
assert conn.pow(2,3) == 8
assert conn.os.path.join("a", "b") == "a/b"
version = conn.reval("lambda: Globals.version")
except:
sys.stderr.write("Server tests failed\n")
raise
if not version == Globals.version:
print """Server may work, but there is a version mismatch:
Local version: %s
Remote version: %s""" % (Globals.version, version)
else: print "Server OK"
from log import *
from rpath import *
from connection import *
import Globals, FilenameMapping
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
#!/usr/bin/env python
import sys, os
from distutils.core import setup, Extension
assert len(sys.argv) == 1
sys.argv.append("build")
setup(name="CModule",
version="0.9.0",
description="rdiff-backup's C component",
ext_modules=[Extension("C", ["cmodule.c"]),
Extension("_librsync",
["_librsyncmodule.c"],
libraries=["rsync"])])
assert not os.system("mv build/lib.linux-i686-2.2/C.so .")
assert not os.system("mv build/lib.linux-i686-2.2/_librsync.so .")
assert not os.system("rm -rf build")
This diff is collapsed.
This diff is collapsed.
from __future__ import generators
execfile("manage.py")
#######################################################################
#
# filelist - Some routines that help with operations over files listed
# in standard input instead of over whole directories.
#
class FilelistError(Exception): pass
class Filelist:
"""Many of these methods have analogs in highlevel.py"""
def File2Iter(fp, baserp):
"""Convert file obj with one pathname per line into rpiter
Closes fp when done. Given files are added to baserp.
"""
while 1:
line = fp.readline()
if not line: break
if line[-1] == "\n": line = line[:-1] # strip trailing newline
if not line: continue # skip blank lines
elif line[0] == "/": raise FilelistError(
"Read in absolute file name %s." % line)
yield baserp.append(line)
assert not fp.close(), "Error closing filelist fp"
def Mirror(src_rpath, dest_rpath, rpiter):
"""Copy files in fileiter from src_rpath to dest_rpath"""
sigiter = dest_rpath.conn.Filelist.get_sigs(dest_rpath, rpiter)
diffiter = Filelist.get_diffs(src_rpath, sigiter)
dest_rpath.conn.Filelist.patch(dest_rpath, diffiter)
dest_rpath.setdata()
def Mirror_and_increment(src_rpath, dest_rpath, inc_rpath):
"""Mirror + put increment in tree based at inc_rpath"""
sigiter = dest_rpath.conn.Filelist.get_sigs(dest_rpath, rpiter)
diffiter = Filelist.get_diffs(src_rpath, sigiter)
dest_rpath.conn.Filelist.patch_and_increment(dest_rpath, diffiter,
inc_rpath)
dest_rpath.setdata()
def get_sigs(dest_rpbase, rpiter):
"""Get signatures of file analogs in rpiter
This is meant to be run on the destination side. Only the
extention part of the rps in rpiter will be used; the base is
ignored.
"""
def dest_iter(src_iter):
for src_rp in src_iter: yield dest_rpbase.new_index(src_rp.index)
return RORPIter.Signatures(dest_iter())
def get_diffs(src_rpbase, sigiter):
"""Get diffs based on sigiter and files in src_rpbase
This should be run on the local side.
"""
for sig_rorp in sigiter:
new_rp = src_rpbase.new_index(sig_rorp.index)
yield RORPIter.diffonce(sig_rorp, new_rp)
def patch(dest_rpbase, diffiter):
"""Process diffs in diffiter and update files in dest_rbpase.
Run remotely.
"""
for diff_rorp in diffiter:
basisrp = dest_rpbase.new_index(diff_rorp.index)
if basisrp.lstat(): Filelist.make_subdirs(basisrp)
Log("Processing %s" % basisrp.path, 7)
RORPIter.patchonce(dest_rpbase, basisrp, diff_rorp)
def patch_and_increment(dest_rpbase, diffiter, inc_rpbase):
"""Apply diffs in diffiter to dest_rpbase, and increment to inc_rpbase
Also to be run remotely.
"""
for diff_rorp in diffiter:
basisrp = dest_rpbase.new_index(diff_rorp.index)
if diff_rorp.lstat(): Filelist.make_subdirs(basisrp)
Log("Processing %s" % basisrp.path, 7)
# XXX This isn't done yet...
def make_subdirs(rpath):
"""Make sure that all the directories under the rpath exist
This function doesn't try to get the permissions right on the
underlying directories, just do the minimum to make sure the
file can be created.
"""
dirname = rpath.dirsplit()[0]
if dirname == '.' or dirname == '': return
dir_rp = RPath(rpath.conn, dirname)
Filelist.make_subdirs(dir_rp)
if not dir_rp.lstat(): dir_rp.mkdir()
MakeStatic(Filelist)
#!/usr/bin/env python
#
# rdiff-backup -- Mirror files while keeping incremental changes
# Version 0.8.0 released June 14, 2002
# Copyright (C) 2001, 2002 Ben Escoto <bescoto@stanford.edu>
#
# This program is licensed under the GNU General Public License (GPL).
# Distributions of rdiff-backup usually include a copy of the GPL in a
# file called COPYING. The GPL is also available online at
# http://www.gnu.org/copyleft/gpl.html.
#
# See http://www.stanford.edu/~bescoto/rdiff-backup for more
# information. Please send mail to me or the mailing list if you find
# bugs or have any suggestions.
from __future__ import nested_scopes, generators
import os, stat, time, sys, getopt, re, cPickle, types, shutil, sha, marshal, traceback, popen2, tempfile, gzip, UserList, errno, signal
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
#!/usr/bin/env python
"""Used to emulate a remote connection by changing directories.
If given an argument, will change to that directory, and then start
the server. Otherwise will start the server without a chdir.
"""
import os, sys
if len(sys.argv) > 1: os.chdir(sys.argv[1])
#PipeConnection(sys.stdin, sys.stdout).Server()
#os.system("/home/ben/prog/python/rdiff-backup/rdiff-backup --server")
os.system("/home/ben/prog/rdiff-backup/testing/server.py /home/ben/prog/rdiff-backup/rdiff_backup")
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment