anteater/src/project_scan.py

   1 #!/usr/bin/env python
   2 # -*- coding: utf-8 -*-
   3 ##############################################################################
   4 # Copyright (c) 2017 Luke Hinds <lhinds@redhat.com>, Red Hat
   5 #
   6 # All rights reserved. This program and the accompanying materials
   7 # are made available under the terms of the Apache License, Version 2.0
   8 # which accompanies this distribution, and is available at
   9 # http://www.apache.org/licenses/LICENSE-2.0
  10 ##############################################################################
  11
  12 """
  13     Accepts the --path argument and iterates the root directory using os.walk
  14     If a file is a binary, or contains a blacklisted string. If any violations
  15     are found, the script adds the violation to a log file.
  16 """
  17
  18 from __future__ import division, print_function, absolute_import
  19 import hashlib
  20 import six.moves.configparser
  21 import os
  22 import re
  23 import logging
  24 from binaryornot.check import is_binary
  25
  26 from . import get_lists
  27
  28 logger = logging.getLogger(__name__)
  29 config = six.moves.configparser.RawConfigParser()
  30 config.read('anteater.conf')
  31 reports_dir = config.get('config', 'reports_dir')
  32 master_list = config.get('config', 'master_list')
  33 ignore_dirs = ['.git']
  34 hasher = hashlib.sha256()
  35
  36
  37 def prepare_project(project, project_dir):
  38     """ Generates blacklists / whitelists and calls main functions """
  39
  40     # Get Various Lists / Project Waivers
  41     lists = get_lists.GetLists()
  42
  43     # Get binary white list
  44     binary_list = lists.binary_list(project)
  45
  46     # Get file name black list and project waivers
  47     file_audit_list, file_audit_project_list = lists.file_audit_list(project)
  48
  49     # Get file content black list and project waivers
  50     master_list, project_list = lists.file_content_list(project)
  51
  52     # Get Licence Lists
  53     licence_ext = lists.licence_extensions()
  54     licence_ignore = lists.licence_ignore()
  55
  56     # Perform rudimentary scans
  57     scan_file(project_dir, project, binary_list,file_audit_list,
  58               file_audit_project_list, master_list,
  59               project_list)
  60
  61     # Perform licence header checks
  62     licence_check(licence_ext, licence_ignore, project, project_dir)
  63     licence_root_check(project_dir, project)
  64
  65
  66 def scan_file(project_dir, project, binary_list, file_audit_list,
  67               file_audit_project_list, master_list,
  68               project_list):
  69     """Searches for banned strings and files that are listed """
  70     for root, dirs, files in os.walk(project_dir):
  71         # Filter out ignored directories from list.
  72         dirs[:] = [d for d in dirs if d not in ignore_dirs]
  73         for items in files:
  74             full_path = os.path.join(root, items)
  75             # Check for Blacklisted file names
  76             if file_audit_list.search(full_path) and not \
  77                     file_audit_project_list.search(full_path):
  78                 match = file_audit_list.search(full_path)
  79                 logger.error('Blacklisted filename: %s', full_path)
  80                 logger.error('Matched String: %s', match.group())
  81                 with open(reports_dir + "file-names_" + project + ".log",
  82                           "a") as gate_report:
  83                             gate_report. \
  84                                 write('Blacklisted filename: {0}\n'.
  85                                       format(full_path))
  86                             gate_report. \
  87                                 write('Matched String: {0}'.
  88                                       format(match.group()))
  89
  90             if not is_binary(full_path):
  91                 try:
  92                     fo = open(full_path, 'r')
  93                     lines = fo.readlines()
  94                 except IOError:
  95                     logger.error('%s does not exist', full_path)
  96
  97                 for line in lines:
  98                     # Check for sensitive content in project files
  99                     for key, value in master_list.iteritems():
 100                         regex = value['regex']
 101                         desc = value['desc']
 102                         if re.search(regex, line) and not re.search(project_list, line):
 103                             logger.error('File contains violation: %s', full_path)
 104                             logger.error('Flagged Content: %s', line.rstrip())
 105                             logger.error('Matched Regular Exp: %s', regex)
 106                             logger.error('Rationale: %s', desc.rstrip())
 107                         with open(reports_dir + "contents-" + project + ".log",
 108                                   "a") \
 109                                 as gate_report:
 110                                     gate_report. \
 111                                         write('File contains violation: {0}\n'.
 112                                               format(full_path))
 113                                     gate_report. \
 114                                         write('Flagged Content: {0}'.
 115                                               format(line))
 116                                     gate_report. \
 117                                         write('Matched Regular Exp: {0}'.
 118                                               format(regex))
 119                                     gate_report. \
 120                                         write('Rationale: {0}\n'.
 121                                               format(desc.rstrip()))
 122             else:
 123                 # Check if Binary is whitelisted
 124                 hashlist = get_lists.GetLists()
 125                 binary_hash = hashlist.binary_hash(project, full_path)
 126                 if not binary_list.search(full_path):
 127                     with open(full_path, 'rb') as afile:
 128                         buf = afile.read()
 129                         hasher.update(buf)
 130                     if hasher.hexdigest() in binary_hash:
 131                         logger.info('Found matching file hash for file: %s',
 132                                     full_path)
 133                     else:
 134                         logger.error('Non Whitelisted Binary file: %s',
 135                                      full_path)
 136                         logger.error('Please submit patch with this hash: %s',
 137                                      hasher.hexdigest())
 138                         with open(reports_dir + "binaries-" + project + ".log",
 139                                   "a") \
 140                                 as gate_report:
 141                             gate_report.write('Non Whitelisted Binary: {0}\n'.
 142                                               format(full_path))
 143
 144
 145 def licence_root_check(project_dir, project):
 146     if os.path.isfile(project_dir + '/LICENSE'):
 147         logger.info('LICENSE file present in: %s', project_dir)
 148     else:
 149         logger.error('LICENSE file missing in: %s', project_dir)
 150         with open(reports_dir + "licence-" + project + ".log",
 151                   "a") \
 152                 as gate_report:
 153             gate_report.write('LICENSE file missing in: {0}\n'.
 154                               format(project_dir))
 155
 156
 157 def licence_check(licence_ext, licence_ignore, project, project_dir):
 158     """ Peform basic checks for the presence of licence strings """
 159     for root, dirs, files in os.walk(project_dir):
 160         dirs[:] = [d for d in dirs if d not in ignore_dirs]
 161         for file in files:
 162             if file.endswith(tuple(licence_ext)) \
 163                     and file not in licence_ignore:
 164                 full_path = os.path.join(root, file)
 165                 if not is_binary(full_path):
 166                     fo = open(full_path, 'r')
 167                     content = fo.read()
 168                     # Note: Hardcoded use of 'copyright' & 'spdx' is the result
 169                     # of a decision made at 2017 plugfest to limit searches to
 170                     # just these two strings.
 171                     patterns = ['copyright', 'spdx',
 172                                 'http://creativecommons.org/licenses/by/4.0']
 173                     if any(i in content.lower() for i in patterns):
 174                         logger.info('Licence string present: %s', full_path)
 175                     else:
 176                         logger.error('Licence header missing: %s', full_path)
 177                         with open(reports_dir + "licence-" + project + ".log",
 178                                   "a") \
 179                                 as gate_report:
 180                             gate_report.write('Licence header missing: {0}\n'.
 181                                               format(full_path))