vmextract.py

   1 #! /usr/bin/python
   2 # -*- coding: utf-8 -*-
   3 #
   4 #  Copyright 2015 Neil Williams <codehelp@debian.org>
   5 #
   6 # This program is free software: you can redistribute it and/or modify
   7 # it under the terms of the GNU General Public License as published by
   8 # the Free Software Foundation, either version 3 of the License, or
   9 # (at your option) any later version.
  10 #
  11 # This program is distributed in the hope that it will be useful,
  12 # but WITHOUT ANY WARRANTY; without even the implied warranty of
  13 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  14 # GNU General Public License for more details.
  15 #
  16 # You should have received a copy of the GNU General Public License
  17 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
  18
  19 import os
  20 import sys
  21 import cliapp
  22 import guestfs
  23 import tarfile
  24 import logging
  25
  26 __version__ = '0.1'
  27 __desc__ = "Helper to mount an image read-only and extract files" + \
  28            " or directories."
  29
  30
  31 # pylint: disable=missing-docstring
  32
  33
  34 class VmExtract(cliapp.Application):  # pylint: disable=too-many-public-methods
  35     """
  36     Support for extracting useful content from VM images.
  37     For example, to assist in validation.
  38     """
  39
  40     def __init__(
  41             self, progname=None,
  42             version=__version__, description=__desc__, epilog=None):
  43         super(VmExtract, self).__init__(
  44             progname, version,
  45             description, epilog)
  46         self.guest_os = None
  47         self.mps = []
  48
  49     def add_settings(self):
  50         self.settings.boolean(
  51             ['verbose'], 'report what is going on')
  52         self.settings.string(
  53             ['image'], 'image to read', metavar='FILE')
  54         self.settings.string(
  55             ['directory'], 'directory to extract as a tarball.')
  56         self.settings.string_list(
  57             ['path'], 'path to the filename to extract - can repeat.')
  58         self.settings.boolean(
  59             ['boot'], 'mount the boot partition as well as root')
  60         self.settings.string(
  61             ['filename'],
  62             'name of tarball containing the extracted directory',
  63             default='vmextract.tgz',
  64             metavar='FILE')
  65
  66     # pylint: disable=too-many-branches,too-many-statements
  67     def process_args(self, args):
  68
  69         if not self.settings['image']:
  70             raise cliapp.AppException(
  71                 'You must give an image to read')
  72         if not self.settings['directory'] and not self.settings['path']:
  73             raise cliapp.AppException(
  74                 'You must provide either a filename or directory '
  75                 'to extract')
  76
  77         try:
  78             self.prepare()
  79             self.mount_root()
  80             if self.settings['boot']:
  81                 self.mount_boot()
  82             if self.settings['directory']:
  83                 self.extract_directory()
  84             elif self.settings['path']:
  85                 for path in self.settings['path']:
  86                     self.download(path)
  87         except BaseException as exc:
  88             self.message('EEEK! Something bad happened... %s' % exc)
  89             sys.exit(1)
  90
  91     def message(self, msg):
  92         logging.info(msg)
  93         if self.settings['verbose']:
  94             print msg
  95
  96     def prepare(self):
  97         """
  98         Initialise guestfs
  99         """
 100         self.message("Preparing %s" % self.settings['image'])
 101         self.guest_os = guestfs.GuestFS(python_return_dict=True)
 102         self.guest_os.add_drive_opts(
 103             self.settings['image'],
 104             format="raw",
 105             readonly=1)
 106         # ensure launch is only called once per run
 107         self.guest_os.launch()
 108         drives = self.guest_os.inspect_os()
 109         self.mps = self.guest_os.inspect_get_mountpoints(drives[0])
 110
 111     def download(self, path):
 112         """
 113         Copy a single file out of the image
 114         If a filename is not specified, use the basename of the original.
 115         """
 116         filename = os.path.basename(path)
 117         self.message(
 118             "Extracting %s as %s" % (path, filename))
 119         self.guest_os.download(path, filename)
 120         if not os.path.exists(filename):
 121             return RuntimeError("Download failed")
 122
 123     def mount_root(self):
 124         """
 125         Mounts the root partition to /
 126         """
 127         root = [part for part in self.mps if part == '/'][0]
 128         if not root:
 129             raise RuntimeError("Unable to identify root partition")
 130         self.guest_os.mount_ro(self.mps[root], '/')
 131
 132     def mount_boot(self):
 133         """
 134         Mounts the /boot partition to a new /boot mountpoint
 135         """
 136         boot = [part for part in self.mps if part == '/boot'][0]
 137         if not boot:
 138             raise RuntimeError("Unable to identify boot partition")
 139         if not self.guest_os.is_dir('/boot'):
 140             self.guest_os.mkmountpoint('/boot')
 141         self.guest_os.mount_ro(self.mps[boot], '/boot')
 142
 143     def extract_directory(self):
 144         """
 145         Create a tarball of a complete directory existing in the image.
 146         """
 147         if not self.settings['filename']:
 148             self.settings['filename'] = 'vmextract.tgz'
 149         self.guest_os.tar_out(
 150             self.settings['directory'],
 151             self.settings['filename'], compress='gzip')
 152         if not tarfile.is_tarfile(self.settings['filename']):
 153             raise RuntimeError("Extraction failed")
 154
 155
 156 def main():
 157     VmExtract(version=__version__).run()
 158     return 0
 159
 160
 161 if __name__ == '__main__':
 162     main()