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 anteater.utils.anteater_logger as antlog
  24 from binaryornot.check import is_binary
  25
  26 from . import get_lists
  27
  28 logger = antlog.Logger(__name__).getLogger()
  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     file_content_list, project_content_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, file_content_list,
  59               project_content_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, file_content_list,
  68               project_content_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: {0}'.
  80                              format(full_path))
  81                 logger.error('Matched String: {0}'.
  82                              format(match.group()))
  83                 with open(reports_dir + "file-names_" + project + ".log",
  84                           "a") as gate_report:
  85                             gate_report. \
  86                                 write('Blacklisted filename: {0}\n'.
  87                                       format(full_path))
  88                             gate_report. \
  89                                 write('Matched String: {0}'.
  90                                       format(match.group()))
  91
  92             if not is_binary(full_path):
  93                 fo = open(full_path, 'r')
  94                 lines = fo.readlines()
  95                 for line in lines:
  96                     # Check for sensitive content in project files
  97                     if file_content_list.search(line) and not \
  98                             project_content_list.search(line):
  99                         match = file_content_list.search(line)
 100                         logger.error('File contains violation: {0}'.
 101                                      format(full_path))
 102                         logger.error('Flagged Content: {0}'.
 103                                      format(line.rstrip()))
 104                         logger.error('Matched String: {0}'.
 105                                      format(match.group()))
 106                         with open(reports_dir + "contents-" + project + ".log",
 107                                   "a") \
 108                                 as gate_report:
 109                                     gate_report. \
 110                                         write('File contains violation: {0}\n'.
 111                                               format(full_path))
 112                                     gate_report. \
 113                                         write('Flagged Content: {0}'.
 114                                               format(line))
 115                                     gate_report. \
 116                                         write('Matched String: {0}\n'.
 117                                               format(match.group()))
 118             else:
 119                 # Check if Binary is whitelisted
 120                 hashlist = get_lists.GetLists()
 121                 binary_hash = hashlist.binary_hash(project, full_path)
 122                 if not binary_list.search(full_path):
 123                     with open(full_path, 'rb') as afile:
 124                         buf = afile.read()
 125                         hasher.update(buf)
 126                     if hasher.hexdigest() in binary_hash:
 127                         logger.info('Found matching file hash for file: {0}'.
 128                                     format(full_path))
 129                     else:
 130                         logger.error('Non Whitelisted Binary file: {0}'.
 131                                      format(full_path))
 132                         logger.error('Please submit patch with this hash: {0}'.
 133                                      format(hasher.hexdigest()))
 134                         with open(reports_dir + "binaries-" + project + ".log",
 135                                   "a") \
 136                                 as gate_report:
 137                             gate_report.write('Non Whitelisted Binary: {0}\n'.
 138                                               format(full_path))
 139
 140
 141 def licence_root_check(project_dir, project):
 142     if os.path.isfile(project_dir + '/LICENSE'):
 143         logger.info('LICENSE file present in: {0}'.
 144                     format(project_dir))
 145     else:
 146         logger.error('LICENSE file missing in: {0}'.
 147                      format(project_dir))
 148         with open(reports_dir + "licence-" + project + ".log",
 149                   "a") \
 150                 as gate_report:
 151             gate_report.write('LICENSE file missing in: {0}\n'.
 152                               format(project_dir))
 153
 154
 155 def licence_check(licence_ext, licence_ignore, project, project_dir):
 156     """ Peform basic checks for the presence of licence strings """
 157     for root, dirs, files in os.walk(project_dir):
 158         dirs[:] = [d for d in dirs if d not in ignore_dirs]
 159         for file in files:
 160             if file.endswith(tuple(licence_ext)) \
 161                     and file not in licence_ignore:
 162                 full_path = os.path.join(root, file)
 163                 if not is_binary(full_path):
 164                     fo = open(full_path, 'r')
 165                     content = fo.read()
 166                     # Note: Hardcoded use of 'copyright' & 'spdx' is the result
 167                     # of a decision made at 2017 plugfest to limit searches to
 168                     # just these two strings.
 169                     if re.search("copyright", content, re.IGNORECASE):
 170                         logger.info('Licence string present: {0}'.
 171                                     format(full_path))
 172                     elif re.search("spdx", content, re.IGNORECASE):
 173                         logger.info('Licence string present: {0}'.
 174                                     format(full_path))
 175                     else:
 176                         logger.error('Licence header missing: {0}'.
 177                                      format(full_path))
 178                         with open(reports_dir + "licence-" + project + ".log",
 179                                   "a") \
 180                                 as gate_report:
 181                             gate_report.write('Licence header missing: {0}\n'.
 182                                               format(full_path))