anteater/src/patch_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 --patchset argument and iterates through each line of the
  14     patchset file to perform various checks such as if the file is a binary, or
  15     contains a blacklisted string. If any violations are found, the script
  16     exits with code 1 and logs the violation(s) found.
  17 """
  18
  19 from __future__ import division, print_function, absolute_import
  20 from binaryornot.check import is_binary
  21 import logging
  22 import hashlib
  23 import six.moves.configparser
  24 import sys
  25 import re
  26
  27 from . import get_lists
  28
  29 logger = logging.getLogger(__name__)
  30 config = six.moves.configparser.RawConfigParser()
  31 config.read('anteater.conf')
  32 reports_dir = config.get('config', 'reports_dir')
  33 failure = False
  34 hasher = hashlib.sha256()
  35
  36
  37 def prepare_patchset(project, patchset):
  38     """ Create black/white lists and default / project waivers
  39         and iterates over patchset file """
  40
  41     # Get Various Lists / Project Waivers
  42     lists = get_lists.GetLists()
  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, \
  51         file_content_project_list = lists.file_content_list(project)
  52
  53     # Get Licence Lists
  54     licence_ext = lists.licence_extensions()
  55     licence_ignore = lists.licence_ignore()
  56
  57     # Open patch set to get file list
  58     try:
  59         fo = open(patchset, 'r')
  60         lines = fo.readlines()
  61     except IOError:
  62         logger.error('%s does not exist', patchset)
  63         sys.exit(1)
  64
  65     for line in lines:
  66         patch_file = line.strip('\n')
  67         # Perform binary and file / content checks
  68         scan_patch(project, patch_file, binary_list,
  69                    file_audit_list, file_audit_project_list,
  70                    file_content_list, file_content_project_list, licence_ext,
  71                    licence_ignore)
  72
  73     # Process each file in patch set using waivers generated above
  74     # Process final result
  75     process_failure()
  76
  77
  78 def scan_patch(project, patch_file, binary_list, file_audit_list,
  79                file_audit_project_list, file_content_list,
  80                file_content_project_list, licence_ext, licence_ignore):
  81     """ Scan actions for each commited file in patch set """
  82     global failure
  83     if is_binary(patch_file):
  84         hashlist = get_lists.GetLists()
  85         binary_hash = hashlist.binary_hash(project, patch_file)
  86         if not binary_list.search(patch_file):
  87             with open(patch_file, 'rb') as afile:
  88                 buf = afile.read()
  89                 hasher.update(buf)
  90             if hasher.hexdigest() in binary_hash:
  91                 logger.info('Found matching file hash for file: %s',
  92                             patch_file)
  93             else:
  94                 logger.error('Non Whitelisted Binary file: %s',
  95                              patch_file)
  96                 logger.error('Submit patch with the following hash: %s',
  97                              hasher.hexdigest())
  98             failure = True
  99             with open(reports_dir + "binaries-" + project + ".log", "a") \
 100                     as gate_report:
 101                 gate_report.write('Non Whitelisted Binary file: {0}\n'.
 102                                   format(patch_file))
 103     else:
 104         # Check file names / extensions
 105         if file_audit_list.search(patch_file) and not \
 106                     file_audit_project_list.search(patch_file):
 107             match = file_audit_list.search(patch_file)
 108             logger.error('Blacklisted file: %s', patch_file)
 109             logger.error('Matched String: %s', match.group())
 110             failure = True
 111             with open(reports_dir + "file-names_" + project + ".log", "a") \
 112                     as gate_report:
 113                 gate_report.write('Blacklisted file: {0}\n'.
 114                                   format(patch_file))
 115                 gate_report.write('Matched String: {0}'.
 116                                   format(match.group()))
 117
 118         # Open file to check for blacklisted content
 119         fo = open(patch_file, 'r')
 120         lines = fo.readlines()
 121
 122         for line in lines:
 123             if file_content_list.search(line) and not \
 124                     file_content_project_list.search(line):
 125                 match = file_content_list.search(line)
 126                 logger.error('File contains violation: %s', patch_file)
 127                 logger.error('Flagged Content: %s', line.rstrip())
 128                 logger.error('Matched String: %s', match.group())
 129                 failure = True
 130                 with open(reports_dir + "contents_" + project + ".log",
 131                           "a") as gate_report:
 132                     gate_report.write('File contains violation: {0}\n'.
 133                                       format(patch_file))
 134                     gate_report.write('Flagged Content: {0}'.
 135                                       format(line))
 136                     gate_report.write('Matched String: {0}\n'.
 137                                       format(match.group()))
 138
 139         # Run license check
 140         licence_check(project, licence_ext, licence_ignore, patch_file)
 141
 142
 143 def licence_check(project, licence_ext,
 144                   licence_ignore, patch_file):
 145     """ Performs licence checks """
 146     global failure
 147     if patch_file.endswith(tuple(licence_ext)) \
 148             and patch_file not in licence_ignore:
 149         fo = open(patch_file, 'r')
 150         content = fo.read()
 151         # Note: Hardcoded use of 'copyright' & 'spdx' is the result
 152         # of a decision made at 2017 plugfest to limit searches to
 153         # just these two strings.
 154         patterns = ['copyright', 'spdx',
 155                     'http://creativecommons.org/licenses/by/4.0']
 156         if any(i in content.lower() for i in patterns):
 157             logger.info('Contains needed Licence string: %s', patch_file)
 158         else:
 159             logger.error('Licence header missing in file: %s', patch_file)
 160             failure = True
 161             with open(reports_dir + "licence-" + project + ".log", "a") \
 162                     as gate_report:
 163                 gate_report.write('Licence header missing in file: {0}\n'.
 164                                   format(patch_file))
 165
 166
 167 def process_failure():
 168     """ If any scan operations register a failure, sys.exit(1) is called
 169         to allow jjb to register a failure"""
 170     if failure:
 171         logger.error('Please visit: https://wiki.opnfv.org/x/5oey')
 172         sys.exit(1)