tools/make-enterprises.py

   1 #!/usr/bin/env python3
   2 # create the enterprises.c file from
   3 # https://www.iana.org/assignments/enterprise-numbers/enterprise-numbers
   4 # or an offline copy
   5 #
   6 # Copyright 2022 by Moshe Kaplan
   7 # Based on make-sminmpec.pl by Gerald Combs
   8 #
   9 # Wireshark - Network traffic analyzer
  10 # By Gerald Combs <gerald@wireshark.org>
  11 # Copyright 2004 Gerald Combs
  12 #
  13 # SPDX-License-Identifier: GPL-2.0-or-later
  14
  15 import os
  16 import argparse
  17 import re
  18 import urllib.request
  19
  20
  21 ENTERPRISES_CFILE = os.path.join('epan', 'enterprises.c')
  22
  23 ENTERPRISE_NUMBERS_URL = "https://www.iana.org/assignments/enterprise-numbers/enterprise-numbers"
  24
  25 DECIMAL_PATTERN = r"^(\d+)"
  26 # up to three spaces because of formatting errors in the source
  27 ORGANIZATION_PATTERN = r"^   ?(\S.*)"
  28 FORMERLY_PATTERN = r" \(((formerly|previously) .*)\)"
  29
  30
  31 LOOKUP_FUNCTION = r"""
  32 const char* global_enterprises_lookup(uint32_t value)
  33 {
  34     if (value >= array_length(table)) {
  35         return NULL;
  36     }
  37     return table[value];
  38 }
  39 """
  40
  41 DUMP_FUNCTION = r"""
  42 void global_enterprises_dump(FILE *fp)
  43 {
  44     for (size_t idx = 0; idx < array_length(table); idx++) {
  45         if (table[idx] != NULL) {
  46             fprintf(fp, "%zu\t%s\n", idx, table[idx]);
  47         }
  48     }
  49 }
  50 """
  51
  52 # This intermediate format is no longer written to a file - returned as string
  53 def generate_enterprise_entries(file_content):
  54     # We only care about the "Decimal" and "Organization",
  55     # not the contact or email
  56     org_lines = []
  57     last_updated = ""
  58     end_seen = False
  59     for line in file_content.splitlines():
  60         decimal_match = re.match(DECIMAL_PATTERN, line)
  61         if decimal_match:
  62             decimal = decimal_match.group(0)
  63         elif re.match(ORGANIZATION_PATTERN, line):
  64             organization = line.strip()
  65             if organization.lower() == "unassigned":
  66                 continue
  67             organization = re.sub(FORMERLY_PATTERN, r"\t# \1", organization)
  68             org_lines += [decimal + "\t" + organization]
  69         elif "last updated" in line.lower():
  70             last_updated = line
  71         elif "end of document" in line.lower():
  72             end_seen = True
  73
  74     if not end_seen:
  75         raise Exception('"End of Document" not found. Truncated source file?')
  76
  77     last_updated_line = "/* " + last_updated + " */\n\n"
  78     output = "\n".join(org_lines) + "\n"
  79     return (output,last_updated_line)
  80
  81 class CFile:
  82     def __init__(self, filename, last_updated_line):
  83         self.filename = filename
  84         self.f = open(filename, 'w')
  85         self.mappings = {}
  86         self.highest_num = 0
  87
  88         # Write file header
  89         self.f.write('/* ' + os.path.basename(self.filename) + '\n')
  90         self.f.write(' *\n')
  91         self.f.write(' * Wireshark - Network traffic analyzer\n')
  92         self.f.write(' * By Gerald Combs <gerald@wireshark.org>\n')
  93         self.f.write(' * Copyright 1998 Gerald Combs\n')
  94         self.f.write(' *\n')
  95         self.f.write(' * Do not edit - this file is automatically generated\n')
  96         self.f.write(' * SPDX-License-Identifier: GPL-2.0-or-later\n')
  97         self.f.write(' */\n\n')
  98         self.f.write(last_updated_line)
  99
 100         # Include header files
 101         self.f.write('#include "config.h"\n\n')
 102         self.f.write('#include <stddef.h>\n')
 103         self.f.write('#include <wsutil/array.h>\n')
 104         self.f.write('#include "enterprises.h"\n')
 105         self.f.write('\n\n')
 106
 107     def __del__(self):
 108         # Write static table
 109         self.f.write('static const char * const table[] =\n')
 110         self.f.write('{\n')
 111         # Largest index
 112         # Entries (read from dict)
 113         for n in range(0, self.highest_num+1):
 114             if n not in self.mappings:
 115                 # There are some gaps, write a NULL entry so can lookup by index
 116                 line = '    NULL'
 117             else:
 118                 line = '    "' + self.mappings[n] + '"'
 119             # Add comma.
 120             if n < self.highest_num:
 121                 line += ','
 122             # Add number as aligned comment.
 123             line += ' '*(90-len(line)) + '// ' + str(n)
 124
 125             self.f.write(line+'\n')
 126
 127         # End of array
 128         self.f.write('};\n')
 129         print('Re-generated', self.filename)
 130
 131         # Lookup function
 132         self.f.write(LOOKUP_FUNCTION)
 133
 134         # Dump function
 135         self.f.write(DUMP_FUNCTION)
 136
 137     # Add an individual mapping to the function
 138     def addMapping(self, num, name):
 139         # Handle some escapings
 140         name = name.replace('\\', '\\\\')
 141         name = name.replace('"', '""')
 142
 143         # Record.
 144         self.mappings[num] = name
 145         self.highest_num = num if num>self.highest_num else self.highest_num
 146
 147
 148
 149 def main():
 150     parser = argparse.ArgumentParser(description="Create the {} file.".format(ENTERPRISES_CFILE))
 151     parser.add_argument('--infile')
 152     parser.add_argument('outfile', nargs='?', default=ENTERPRISES_CFILE)
 153     parsed_args = parser.parse_args()
 154
 155     # Read data from file or webpage
 156     if parsed_args.infile:
 157         with open(parsed_args.infile, encoding='utf-8') as fh:
 158             data = fh.read()
 159     else:
 160         with urllib.request.urlopen(ENTERPRISE_NUMBERS_URL) as f:
 161             if f.status != 200:
 162                 raise Exception("request for " + ENTERPRISE_NUMBERS_URL + " failed with result code " + f.status)
 163             data = f.read().decode('utf-8').replace(u'\u200e', '')
 164
 165     # Find bits we need and generate enterprise entries
 166     enterprises_content,last_updated_line = generate_enterprise_entries(data)
 167
 168     # Now write to a C file the contents (which is faster than parsing the global file at runtime).
 169     c_file = CFile(parsed_args.outfile, last_updated_line)
 170
 171     mapping_re = re.compile(r'^(\d+)\s+(.*)$')
 172     for line in enterprises_content.splitlines():
 173         match = mapping_re.match(line)
 174         if match:
 175             num, name = match.group(1), match.group(2)
 176             # Strip any comments and/or trailing whitespace
 177             idx = name.find('#')
 178             if idx != -1:
 179                 name = name[0:idx]
 180             name = name.rstrip()
 181             # Add
 182             c_file.addMapping(int(num), name)
 183
 184
 185
 186 if __name__ == "__main__":
 187     main()