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 ConfigParser
  20 import os
  21 import re
  22 import anteater.utils.anteater_logger as antlog
  23 import anteater.src.get_lists as get_lists
  24 from binaryornot.check import is_binary
  25
  26 logger = antlog.Logger(__name__).getLogger()
  27 config = ConfigParser.RawConfigParser()
  28 config.read('anteater.conf')
  29 reports_dir = config.get('config', 'reports_dir')
  30 master_list = config.get('config', 'master_list')
  31 ignore_dirs = ['.git']
  32
  33
  34 def prepare_project(project, project_dir):
  35     """ Generates blacklists / whitelists and calls main functions """
  36
  37     # Get Various Lists / Project Waivers
  38     lists = get_lists.GetLists()
  39
  40     # Get binary white list
  41     binary_list, binary_project_list = lists.binary_list(project)
  42
  43     # Get file name black list and project waivers
  44     file_audit_list, file_audit_project_list = lists.file_audit_list(project)
  45
  46     # Get file content black list and project waivers
  47     file_content_list, project_content_list = lists.file_content_list(project)
  48
  49     # Get Licence Lists
  50     licence_ext = lists.licence_extensions()
  51     licence_ignore = lists.licence_ignore()
  52
  53     # Perform rudimentary scans
  54     scan_file(project_dir, project, binary_list, binary_project_list,
  55               file_audit_list, file_audit_project_list, file_content_list,
  56               project_content_list)
  57
  58     # Perform licence header checks
  59     licence_check(licence_ext, licence_ignore, project, project_dir)
  60     licence_root_check(project_dir, project)
  61
  62
  63 def scan_file(project_dir, project, binary_list, binary_project_list,
  64               file_audit_list, file_audit_project_list, file_content_list,
  65               project_content_list):
  66     """Searches for banned strings and files that are listed """
  67     for root, dirs, files in os.walk(project_dir):
  68         # Filter out ignored directories from list.
  69         dirs[:] = [d for d in dirs if d not in ignore_dirs]
  70         for items in files:
  71             full_path = os.path.join(root, items)
  72             # Check for Blacklisted file names
  73             if file_audit_list.search(full_path) and not \
  74                     file_audit_project_list.search(full_path):
  75                 match = file_audit_list.search(full_path)
  76                 logger.error('Blacklisted filename: {0}'.
  77                              format(full_path))
  78                 logger.error('Matched String: {0}'.
  79                              format(match.group()))
  80                 with open(reports_dir + "file-names_" + project + ".log",
  81                           "a") as gate_report:
  82                             gate_report. \
  83                                 write('Blacklisted filename: {0}\n'.
  84                                       format(full_path))
  85                             gate_report. \
  86                                 write('Matched String: {0}'.
  87                                       format(match.group()))
  88
  89             if not is_binary(full_path):
  90                 fo = open(full_path, 'r')
  91                 lines = fo.readlines()
  92                 for line in lines:
  93                     # Check for sensitive content in project files
  94                     if file_content_list.search(line) and not \
  95                             project_content_list.search(line):
  96                         match = file_content_list.search(line)
  97                         logger.error('File contains violation: {0}'.
  98                                      format(full_path))
  99                         logger.error('Flagged Content: {0}'.
 100                                      format(line.rstrip()))
 101                         logger.error('Matched String: {0}'.
 102                                      format(match.group()))
 103                         with open(reports_dir + "contents-" + project + ".log",
 104                                   "a") \
 105                                 as gate_report:
 106                                     gate_report. \
 107                                         write('File contains violation: {0}\n'.
 108                                               format(full_path))
 109                                     gate_report. \
 110                                         write('Flagged Content: {0}'.
 111                                               format(line))
 112                                     gate_report. \
 113                                         write('Matched String: {0}\n'.
 114                                               format(match.group()))
 115             else:
 116                 # Check if Binary is whitelisted
 117                 if not binary_list.search(full_path) \
 118                         and not binary_project_list.search(full_path):
 119                     logger.error('Non Whitelisted Binary: {0}'.
 120                                  format(full_path))
 121                     with open(reports_dir + "binaries-" + project + ".log",
 122                               "a") \
 123                             as gate_report:
 124                         gate_report.write('Non Whitelisted Binary: {0}\n'.
 125                                           format(full_path))
 126
 127
 128 def licence_root_check(project_dir, project):
 129     if os.path.isfile(project_dir + '/LICENSE'):
 130         logger.info('LICENSE file present in: {0}'.
 131                     format(project_dir))
 132     else:
 133         logger.error('LICENSE file missing in: {0}'.
 134                      format(project_dir))
 135         with open(reports_dir + "licence-" + project + ".log",
 136                   "a") \
 137                 as gate_report:
 138             gate_report.write('LICENSE file missing in: {0}\n'.
 139                               format(project_dir))
 140
 141
 142 def licence_check(licence_ext, licence_ignore, project, project_dir):
 143     """ Peform basic checks for the presence of licence strings """
 144     for root, dirs, files in os.walk(project_dir):
 145         dirs[:] = [d for d in dirs if d not in ignore_dirs]
 146         for file in files:
 147             if file.endswith(tuple(licence_ext)) \
 148                     and file not in licence_ignore:
 149                 full_path = os.path.join(root, file)
 150                 if not is_binary(full_path):
 151                     fo = open(full_path, 'r')
 152                     content = fo.read()
 153                     # Note: Hardcoded use of 'copyright' & 'spdx' is the result
 154                     # of a decision made at 2017 plugfest to limit searches to
 155                     # just these two strings.
 156                     if re.search("copyright", content, re.IGNORECASE):
 157                         logger.info('Licence string present: {0}'.
 158                                     format(full_path))
 159                     elif re.search("spdx", content, re.IGNORECASE):
 160                         logger.info('Licence string present: {0}'.
 161                                     format(full_path))
 162                     else:
 163                         logger.error('Licence header missing: {0}'.
 164                                      format(full_path))
 165                         with open(reports_dir + "licence-" + project + ".log",
 166                                   "a") \
 167                                 as gate_report:
 168                             gate_report.write('Licence header missing: {0}\n'.
 169                                               format(full_path))