tools/make-services.py

   1 #!/usr/bin/env python
   2 #
   3 # Parses the CSV version of the IANA Service Name and Transport Protocol Port Number Registry
   4 # and generates a services(5) file.
   5 #
   6 # Wireshark - Network traffic analyzer
   7 # By Gerald Combs <gerald@wireshark.org>
   8 # Copyright 2013 Gerald Combs
   9 #
  10 # This program is free software; you can redistribute it and/or
  11 # modify it under the terms of the GNU General Public License
  12 # as published by the Free Software Foundation; either version 2
  13 # of the License, or (at your option) any later version.
  14 #
  15 # This program is distributed in the hope that it will be useful,
  16 # but WITHOUT ANY WARRANTY; without even the implied warranty of
  17 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  18 # GNU General Public License for more details.
  19 #
  20 # You should have received a copy of the GNU General Public License
  21 # along with this program; if not, write to the Free Software
  22 # Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
  23
  24 iana_svc_url = 'http://www.iana.org/assignments/service-names-port-numbers/service-names-port-numbers.csv'
  25
  26 __doc__ = '''\
  27 Usage: make-services.py [url]
  28
  29 url defaults to
  30     %s
  31 ''' % (iana_svc_url)
  32
  33 import sys
  34 import getopt
  35 import csv
  36 import re
  37
  38 python_version = sys.hexversion >> 16
  39 if python_version < 0x300:
  40     import urllib
  41 else:
  42     import urllib.request, urllib.error, urllib.parse
  43     import codecs
  44
  45 services_file = 'services'
  46
  47 exclude_services = [
  48     '^spr-itunes',
  49     '^spl-itunes',
  50     '^shilp',
  51     ]
  52
  53 exclude_comments = [
  54     'should not be used for discovery purposes',
  55     'NOTE Conflict',
  56 ]
  57
  58 min_body_size = 900000 # Size was ~ 922000 on 2013-08-06
  59
  60 def parse_rows(svc_fd):
  61     lines = []
  62     port_reader = csv.reader(svc_fd)
  63
  64     # Header positions as of 2013-08-06
  65     if python_version < 0x206:
  66         headers = port_reader.next()
  67     else:
  68         headers = next(port_reader)
  69
  70     try:
  71         sn_pos = headers.index('Service Name')
  72     except:
  73         sn_pos = 0
  74     try:
  75         pn_pos = headers.index('Port Number')
  76     except:
  77         pn_pos = 1
  78     try:
  79         tp_pos = headers.index('Transport Protocol')
  80     except:
  81         tp_pos = 2
  82
  83     positions = [sn_pos, pn_pos, tp_pos]
  84     positions.sort()
  85     positions.reverse()
  86
  87     for row in port_reader:
  88         service = row[sn_pos]
  89         port = row[pn_pos]
  90         proto = row[tp_pos]
  91
  92         if len(service) < 1 or len(port) < 1 or len(proto) < 1:
  93             continue
  94
  95         for pos in positions:
  96             del row[pos]
  97         row = filter(None, row)
  98         comment = ' '.join(row)
  99         comment = re.sub('[\n]', '', comment)
 100
 101         if re.search('|'.join(exclude_services), service):
 102             continue
 103         if re.search('|'.join(exclude_comments), comment):
 104             continue
 105
 106         lines.append('%-15s %5s/%s # %s' % (
 107             service,
 108             port,
 109             proto,
 110             comment
 111         ))
 112
 113     return '\n'.join(lines)
 114
 115 def exit_msg(msg=None, status=1):
 116     if msg is not None:
 117         sys.stderr.write(msg + '\n\n')
 118     sys.stderr.write(__doc__ + '\n')
 119     sys.exit(status)
 120
 121 def main(argv):
 122     try:
 123         opts, args = getopt.getopt(argv, "h", ["help"])
 124     except getopt.GetoptError:
 125         exit_msg()
 126     for opt, arg in opts:
 127         if opt in ("-h", "--help"):
 128             exit_msg(None, 0)
 129
 130     if (len(argv) > 0):
 131         svc_url = argv[0]
 132     else:
 133         svc_url = iana_svc_url
 134
 135     try:
 136         if python_version < 0x300:
 137             svc_fd = urllib.urlopen(svc_url)
 138         else:
 139             req = urllib.request.urlopen(svc_url)
 140             svc_fd = codecs.getreader('utf8')(req)
 141     except:
 142         exit_msg('Error opening ' + svc_url)
 143
 144     body = parse_rows(svc_fd)
 145     if len(body) < min_body_size:
 146         exit_msg('Not enough parsed data')
 147
 148     out = open(services_file, 'w')
 149     out.write('''\
 150 # This is a local copy of the IANA port-numbers file.
 151 #
 152 # $Id$
 153 #
 154 # Wireshark uses it to resolve port numbers into human readable
 155 # service names, e.g. TCP port 80 -> http.
 156 #
 157 # It is subject to copyright and being used with IANA's permission:
 158 # http://www.wireshark.org/lists/wireshark-dev/200708/msg00160.html
 159 #
 160 # The original file can be found at:
 161 # %s
 162 #
 163
 164 %s
 165 ''' % (iana_svc_url, body))
 166
 167 if __name__ == "__main__":
 168     sys.exit(main(sys.argv[1:]))