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 File Ignore Lists
  53     file_ignore = lists.file_ignore()
  54
  55     # Get Licence Lists
  56     licence_ext = lists.licence_extensions()
  57     licence_ignore = lists.licence_ignore()
  58
  59     # Perform rudimentary scans
  60     scan_file(project_dir, project, binary_list,file_audit_list,
  61               file_audit_project_list, master_list, file_ignore,
  62               project_list)
  63
  64     # Perform licence header checks
  65     licence_check(licence_ext, licence_ignore, project, project_dir)
  66     licence_root_check(project_dir, project)
  67
  68
  69 def scan_file(project_dir, project, binary_list, file_audit_list,
  70               file_audit_project_list, master_list, file_ignore,
  71               project_list):
  72     """Searches for banned strings and files that are listed """
  73     for root, dirs, files in os.walk(project_dir):
  74         # Filter out ignored directories from list.
  75         dirs[:] = [d for d in dirs if d not in ignore_dirs]
  76         for items in files:
  77             full_path = os.path.join(root, items)
  78             # Check for Blacklisted file names
  79             if file_audit_list.search(full_path) and not \
  80                     file_audit_project_list.search(full_path):
  81                 match = file_audit_list.search(full_path)
  82                 logger.error('Blacklisted filename: %s', full_path)
  83                 logger.error('Matched String: %s', match.group())
  84                 with open(reports_dir + "file-names_" + project + ".log",
  85                           "a") as gate_report:
  86                             gate_report. \
  87                                 write('Blacklisted filename: {0}\n'.
  88                                       format(full_path))
  89                             gate_report. \
  90                                 write('Matched String: {0}'.
  91                                       format(match.group()))
  92
  93                             # Check if Binary is whitelisted
  94             hashlist = get_lists.GetLists()
  95             binary_hash = hashlist.binary_hash(project, full_path)
  96             if is_binary(full_path) and not binary_list.search(full_path):
  97                 with open(full_path, 'rb') as afile:
  98                     buf = afile.read()
  99                     hasher.update(buf)
 100                 if hasher.hexdigest() in binary_hash:
 101                     logger.info('Found matching file hash for file: %s',
 102                                     full_path)
 103                 else:
 104                     logger.error('Non Whitelisted Binary file: %s',
 105                                  full_path)
 106                     logger.error('Please submit patch with this hash: %s',
 107                                  hasher.hexdigest())
 108                     with open(reports_dir + "binaries-" + project + ".log",
 109                               "a") as gate_report:
 110                             gate_report.write('Non Whitelisted Binary: {0}\n'.
 111                                               format(full_path))
 112
 113             else:
 114                 if not items.endswith(tuple(file_ignore)):
 115                     try:
 116                         fo = open(full_path, 'r')
 117                         lines = fo.readlines()
 118                     except IOError:
 119                         logger.error('%s does not exist', full_path)
 120
 121                     for line in lines:
 122                         # Check for sensitive content in project files
 123                         for key, value in master_list.iteritems():
 124                             regex = value['regex']
 125                             desc = value['desc']
 126                             if re.search(regex, line) and not re.search(
 127                                     project_list, line):
 128                                 logger.error('File contains violation: %s',
 129                                              full_path)
 130                                 logger.error('Flagged Content: %s',
 131                                              line.rstrip())
 132                                 logger.error('Matched Regular Exp: %s', regex)
 133                                 logger.error('Rationale: %s', desc.rstrip())
 134                                 with open(reports_dir + "contents-" + project
 135                                                   + ".log", "a") \
 136                                         as gate_report:
 137                                     gate_report. \
 138                                         write('File contains violation: {0}\n'.
 139                                               format(full_path))
 140                                     gate_report. \
 141                                         write('Flagged Content: {0}'.
 142                                               format(line))
 143                                     gate_report. \
 144                                         write('Matched Regular Exp: {0}'.
 145                                               format(regex))
 146                                     gate_report. \
 147                                         write('Rationale: {0}\n'.
 148                                               format(desc.rstrip()))
 149
 150
 151
 152 def licence_root_check(project_dir, project):
 153     if os.path.isfile(project_dir + '/LICENSE'):
 154         logger.info('LICENSE file present in: %s', project_dir)
 155     else:
 156         logger.error('LICENSE file missing in: %s', project_dir)
 157         with open(reports_dir + "licence-" + project + ".log",
 158                   "a") \
 159                 as gate_report:
 160             gate_report.write('LICENSE file missing in: {0}\n'.
 161                               format(project_dir))
 162
 163
 164 def licence_check(licence_ext, licence_ignore, project, project_dir):
 165     """ Peform basic checks for the presence of licence strings """
 166     for root, dirs, files in os.walk(project_dir):
 167         dirs[:] = [d for d in dirs if d not in ignore_dirs]
 168         for file in files:
 169             if file.endswith(tuple(licence_ext)) \
 170                     and file not in licence_ignore:
 171                 full_path = os.path.join(root, file)
 172                 if not is_binary(full_path):
 173                     fo = open(full_path, 'r')
 174                     content = fo.read()
 175                     # Note: Hardcoded use of 'copyright' & 'spdx' is the result
 176                     # of a decision made at 2017 plugfest to limit searches to
 177                     # just these two strings.
 178                     patterns = ['copyright', 'spdx',
 179                                 'http://creativecommons.org/licenses/by/4.0']
 180                     if any(i in content.lower() for i in patterns):
 181                         logger.info('Licence string present: %s', full_path)
 182                     else:
 183                         logger.error('Licence header missing: %s', full_path)
 184                         with open(reports_dir + "licence-" + project + ".log",
 185                                   "a") \
 186                                 as gate_report:
 187                             gate_report.write('Licence header missing: {0}\n'.
 188                                               format(full_path))